Kaggle:Yelp:Restaurant Features: outdoor label classification:: YelpRest_resXY_outdoor
bdanalytics
Date: (Fri) Feb 05, 2016
Introduction:
Data: Source: Training: https://www.kaggle.com/c/yelp-restaurant-photo-classification/download/train.csv.tgz
New: https://www.kaggle.com/c/yelp-restaurant-photo-classification/download/test.csv.tgz
Time period:
Synopsis:
Based on analysis utilizing <> techniques,
Summary of key steps & error improvement stats:
Prediction Accuracy Enhancement Options:
- transform.data chunk:
- derive features from multiple features
- manage.missing.data chunk:
- Not fill missing vars
- Fill missing numerics with a different algorithm
- Fill missing chars with data based on clusters
[](.png)
Potential next steps include:
- Organization:
- Categorize by chunk
- Priority criteria:
- Ease of change
- Impacts report
- Cleans innards
- Bug report
- all chunks:
- at chunk-end rm(!glb_)
- manage.missing.data chunk:
- cleaner way to manage re-splitting of training vs. new entity
- extract.features chunk:
- Add n-grams for glbFeatsText
- “RTextTools”, “tau”, “RWeka”, and “textcat” packages
- Add n-grams for glbFeatsText
- fit.models chunk:
- Prediction accuracy scatter graph:
- Add tiles (raw vs. PCA)
- Use shiny for drop-down of “important” features
Use plot.ly for interactive plots ?
- Change .fit suffix of model metrics to .mdl if it’s data independent (e.g. AIC, Adj.R.Squared - is it truly data independent ?, etc.)
- create a custom model for rpart that has minbucket as a tuning parameter
varImp for randomForest crashes in caret version:6.0.41 -> submit bug report
- Probability handling for multinomials vs. desired binomial outcome
- ROCR currently supports only evaluation of binary classification tasks (version 1.0.7)
extensions toward multiclass classification are scheduled for the next release
- fit.all.training chunk:
- myplot_prediction_classification: displays ‘x’ instead of ‘+’ when there are no prediction errors
- Compare glb_sel_mdl vs. glb_fin_mdl:
- varImp
- Prediction differences (shd be minimal ?)
- Move glb_analytics_diag_plots to mydsutils.R: (+) Easier to debug (-) Too many glb vars used
- Add print(ggplot.petrinet(glb_analytics_pn) + coord_flip()) at the end of every major chunk
- Parameterize glb_analytics_pn
- Move glb_impute_missing_data to mydsutils.R: (-) Too many glb vars used; glb_<>_df reassigned
- Do non-glm methods handle interaction terms ?
- f-score computation for classifiers should be summation across outcomes (not just the desired one ?)
- Add accuracy computation to glb_dmy_mdl in predict.data.new chunk
- Why does splitting fit.data.training.all chunk into separate chunks add an overhead of ~30 secs ? It’s not rbind b/c other chunks have lower elapsed time. Is it the number of plots ?
- Incorporate code chunks in print_sessionInfo
- Test against
- projects in github.com/bdanalytics
- lectures in jhu-datascience track
Analysis:
rm(list = ls())
set.seed(12345)
options(stringsAsFactors = FALSE)
source("~/Dropbox/datascience/R/myscript.R")
source("~/Dropbox/datascience/R/mydsutils.R")## Loading required package: caret## Loading required package: lattice## Loading required package: ggplot2source("~/Dropbox/datascience/R/myplot.R")
source("~/Dropbox/datascience/R/mypetrinet.R")
source("~/Dropbox/datascience/R/myplclust.R")
source("~/Dropbox/datascience/R/mytm.R")
# Gather all package requirements here
suppressPackageStartupMessages(require(doMC))
glbCores <- 6 # of cores on machine - 2
registerDoMC(glbCores)
suppressPackageStartupMessages(require(caret))
require(plyr)## Loading required package: plyrrequire(dplyr)## Loading required package: dplyr##
## Attaching package: 'dplyr'## The following objects are masked from 'package:plyr':
##
## arrange, count, desc, failwith, id, mutate, rename, summarise,
## summarize## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionrequire(knitr)## Loading required package: knitrrequire(stringr)## Loading required package: stringr#source("dbgcaret.R")
#packageVersion("snow")
#require(sos); findFn("cosine", maxPages=2, sortby="MaxScore")
# Analysis control global variables
# Inputs
# url/name = "<pointer>"; if url specifies a zip file, name = "<filename>"
# sep = choose from c(NULL, "\t")
glbObsTrnFile <- list(url = "https://www.kaggle.com/c/yelp-restaurant-photo-classification/download/train.csv.tgz",
name = "train_resXY.csv")
glbObsNewFile <- list(url = "https://www.kaggle.com/c/yelp-restaurant-photo-classification/download/test.csv.tgz",
name = "test_resXY.csv") # default OR
#list(splitSpecs = list(method = NULL #select from c(NULL, "condition", "sample", "copy")
# ,nRatio = 0.3 # > 0 && < 1 if method == "sample"
# ,seed = 123 # any integer or glbObsTrnPartitionSeed if method == "sample"
# ,condition = # or 'is.na(<var>)'; '<var> <condition_operator> <value>'
# )
# )
glbInpMerge <- NULL #: default
# list(fnames = c("<fname1>", "<fname2>")) # files will be concatenated
glb_is_separate_newobs_dataset <- TRUE # or TRUE
glb_split_entity_newobs_datasets <- TRUE # FALSE not supported - use "copy" for glbObsNewFile$splitSpecs$method # select from c(FALSE, TRUE)
glbObsDropCondition <- NULL # : default
# enclose in single-quotes b/c condition might include double qoutes
# use | & ; NOT || &&
# '<condition>'
# 'grepl("^First Draft Video:", glbObsAll$Headline)'
# '(is.na(glbObsAll[, glb_rsp_var_raw]) & grepl("Train", glbObsAll[, glbFeatsId]))'
#nrow(do.call("subset",list(glbObsAll, parse(text=paste0("!(", glbObsDropCondition, ")")))))
glb_obs_repartition_train_condition <- NULL # : default
# "<condition>"
glb_max_fitobs <- NULL # or any integer
glbObsTrnPartitionSeed <- 123 # or any integer
glb_is_regression <- FALSE; glb_is_classification <- !glb_is_regression;
glb_is_binomial <- TRUE # or TRUE or FALSE
glb_rsp_var_raw <- "outdoor"
# for classification, the response variable has to be a factor
glb_rsp_var <- "outdoor.fctr"
# if the response factor is based on numbers/logicals e.g (0/1 OR TRUE/FALSE vs. "A"/"B"),
# or contains spaces (e.g. "Not in Labor Force")
# caret predict(..., type="prob") crashes
glb_map_rsp_raw_to_var <- #NULL
function(raw) {
# return(raw ^ 0.5)
# return(log(raw))
# return(log(1 + raw))
# return(log10(raw))
# return(exp(-raw / 2))
# ret_vals <- rep_len(NA, length(raw)); ret_vals[!is.na(raw)] <- ifelse(raw[!is.na(raw)] == 1, "Y", "N"); return(relevel(as.factor(ret_vals), ref="N"))
ret_vals <- rep_len(NA, length(raw)); ret_vals[!is.na(raw)] <- ifelse(raw[!is.na(raw)] != -1, "Y", "N"); return(relevel(as.factor(ret_vals), ref = "N"))
# as.factor(paste0("B", raw))
# as.factor(gsub(" ", "\\.", raw))
}
#if glb_rsp_var_raw is numeric:
#print(summary(glbObsAll[, glb_rsp_var_raw]))
#glb_map_rsp_raw_to_var(tst <- c(NA, as.numeric(summary(glbObsAll[, glb_rsp_var_raw]))))
#if glb_rsp_var_raw is character:
#print(table(glbObsAll[, glb_rsp_var_raw], useNA = "ifany"))
#print(table(glb_map_rsp_raw_to_var(tst <- glbObsAll[, glb_rsp_var_raw]), useNA = "ifany"))
glb_map_rsp_var_to_raw <- #NULL
function(var) {
# return(var ^ 2.0)
# return(exp(var))
# return(10 ^ var)
# return(-log(var) * 2)
# as.numeric(var)
levels(var)[as.numeric(var)]
# gsub("\\.", " ", levels(var)[as.numeric(var)])
# c("<=50K", " >50K")[as.numeric(var)]
# c(FALSE, TRUE)[as.numeric(var)]
}
#print(table(glb_map_rsp_var_to_raw(glb_map_rsp_raw_to_var(tst)), useNA = "ifany"))
if ((glb_rsp_var != glb_rsp_var_raw) && is.null(glb_map_rsp_raw_to_var))
stop("glb_map_rsp_raw_to_var function expected")
# List info gathered for various columns
# <col_name>: <description>; <notes>
# currently does not handle more than 1 column; consider concatenating multiple columns
# If glbFeatsId == NULL, ".rownames <- as.numeric(row.names())" is the default
glbFeatsId <- "business_id" # choose from c(NULL : default, "<id_feat>")
glbFeatsCategory <- "nImgs.cut.fctr" # choose from c(NULL : default, "<category_feat>")
# User-specified exclusions
glbFeatsExclude <- c(NULL
# Feats that shd be excluded due to known causation by prediction variable
# , "<feat1", "<feat2>"
,"labels"
,"lunch","dinner","reserve","outdoor","expensive","liquor","table","classy","kids"
# Feats that are linear combinations (alias in glm)
# Feature-engineering phase -> start by excluding all features except id & category & work each one in
,"business_id"
,"resXLst","resYLst"
)
if (glb_rsp_var_raw != glb_rsp_var)
glbFeatsExclude <- union(glbFeatsExclude, glb_rsp_var_raw)
glbFeatsInteractionOnly <- list()
#glbFeatsInteractionOnly[["<child_feat>"]] <- "<parent_feat>"
glbFeatsDrop <- c(NULL
# , "<feat1>", "<feat2>"
)
glb_map_vars <- NULL # or c("<var1>", "<var2>")
glb_map_urls <- list();
# glb_map_urls[["<var1>"]] <- "<var1.url>"
glb_assign_pairs_lst <- NULL;
# glb_assign_pairs_lst[["<var1>"]] <- list(from=c(NA),
# to=c("NA.my"))
glb_assign_vars <- names(glb_assign_pairs_lst)
# Derived features; Use this mechanism to cleanse data ??? Cons: Data duplication ???
glbFeatsDerive <- list();
# glbFeatsDerive[["<feat.my.sfx>"]] <- list(
# mapfn = function(<arg1>, <arg2>) { return(function(<arg1>, <arg2>)) }
# , args = c("<arg1>", "<arg2>"))
#myprint_df(data.frame(ImageId = mapfn(glbObsAll$.src, glbObsAll$.pos)))
#data.frame(ImageId = mapfn(glbObsAll$.src, glbObsAll$.pos))[7045:7055, ]
# character
# mapfn = function(Week) { return(substr(Week, 1, 10)) }
# mapfn = function(descriptor) { return(plyr::revalue(descriptor, c(
# "ABANDONED BUILDING" = "OTHER",
# "**" = "**"
# ))) }
# mapfn = function(description) { mod_raw <- description;
# This is here because it does not work if it's in txt_map_filename
# mod_raw <- gsub(paste0(c("\n", "\211", "\235", "\317", "\333"), collapse = "|"), " ", mod_raw)
# Don't parse for "." because of ".com"; use customized gsub for that text
# mod_raw <- gsub("(\\w)(!|\\*|,|-|/)(\\w)", "\\1\\2 \\3", mod_raw);
# Some state acrnoyms need context for separation e.g.
# LA/L.A. could either be "Louisiana" or "LosAngeles"
# modRaw <- gsub("\\bL\\.A\\.( |,|')", "LosAngeles\\1", modRaw);
# OK/O.K. could either be "Oklahoma" or "Okay"
# modRaw <- gsub("\\bACA OK\\b", "ACA OKay", modRaw);
# modRaw <- gsub("\\bNow O\\.K\\.\\b", "Now OKay", modRaw);
# PR/P.R. could either be "PuertoRico" or "Public Relations"
# modRaw <- gsub("\\bP\\.R\\. Campaign", "PublicRelations Campaign", modRaw);
# VA/V.A. could either be "Virginia" or "VeteransAdministration"
# modRaw <- gsub("\\bthe V\\.A\\.\\:", "the VeteranAffairs:", modRaw);
#
# Custom mods
# return(mod_raw) }
# numeric
# Create feature based on record position/id in data
glbFeatsDerive[[".pos"]] <- list(
mapfn = function(.rnorm) { return(1:length(.rnorm)) }
, args = c(".rnorm"))
# glbFeatsDerive[[".pos.y"]] <- list(
# mapfn = function(.rnorm) { return(1:length(.rnorm)) }
# , args = c(".rnorm"))
glbFeatsDerive[["nImgs.log1p"]] <- list(
mapfn = function(nImgs) { return(log1p(nImgs)) }
, args = c("nImgs"))
glbFeatsDerive[["nImgs.root2"]] <- list(
mapfn = function(nImgs) { return(nImgs ^ (1/2)) }
, args = c("nImgs"))
glbFeatsDerive[["nImgs.nexp"]] <- list(
mapfn = function(nImgs) { return(exp(-nImgs)) }
, args = c("nImgs"))
glbFeatsDerive[["resX.min"]] <- list(
mapfn = function(resXLst) { return(sapply(resXLst, function(thsObsFeat)
min(as.numeric(unlist(str_split(thsObsFeat, ",")))))) }
, args = c("resXLst"))
glbFeatsDerive[["resX.max"]] <- list(
mapfn = function(resXLst) { return(sapply(resXLst, function(thsObsFeat)
max(as.numeric(unlist(str_split(thsObsFeat, ",")))))) }
, args = c("resXLst"))
glbFeatsDerive[["resX.mean"]] <- list(
mapfn = function(resXLst) { return(sapply(resXLst, function(thsObsFeat)
mean(as.numeric(unlist(str_split(thsObsFeat, ",")))))) }
, args = c("resXLst"))
glbFeatsDerive[["resX.mad"]] <- list(
mapfn = function(resXLst) { return(sapply(resXLst, function(thsObsFeat)
mad(as.numeric(unlist(str_split(thsObsFeat, ",")))))) }
, args = c("resXLst"))
glbFeatsDerive[["resX.min.log1p"]] <- list(
mapfn = function(resX.min) { return(log1p(resX.min)) }
, args = c("resX.min"))
glbFeatsDerive[["resX.min.root2"]] <- list(
mapfn = function(resX.min) { return(resX.min ^ (1/2)) }
, args = c("resX.min"))
glbFeatsDerive[["resX.min.nexp"]] <- list(
mapfn = function(resX.min) { return(exp(-resX.min)) }
, args = c("resX.min"))
glbFeatsDerive[["resX.max.log1p"]] <- list(
mapfn = function(resX.max) { return(log1p(resX.max)) }
, args = c("resX.max"))
glbFeatsDerive[["resX.max.root2"]] <- list(
mapfn = function(resX.max) { return(resX.max ^ (1/2)) }
, args = c("resX.max"))
glbFeatsDerive[["resX.max.nexp"]] <- list(
mapfn = function(resX.max) { return(exp(-resX.max)) }
, args = c("resX.max"))
glbFeatsDerive[["resX.mean.log1p"]] <- list(
mapfn = function(resX.mean) { return(log1p(resX.mean)) }
, args = c("resX.mean"))
glbFeatsDerive[["resX.mean.root2"]] <- list(
mapfn = function(resX.mean) { return(resX.mean ^ (1/2)) }
, args = c("resX.mean"))
glbFeatsDerive[["resX.mean.nexp"]] <- list(
mapfn = function(resX.mean) { return(exp(-resX.mean)) }
, args = c("resX.mean"))
glbFeatsDerive[["resX.mad.log1p"]] <- list(
mapfn = function(resX.mad) { return(log1p(resX.mad)) }
, args = c("resX.mad"))
glbFeatsDerive[["resX.mad.root2"]] <- list(
mapfn = function(resX.mad) { return(resX.mad ^ (1/2)) }
, args = c("resX.mad"))
glbFeatsDerive[["resX.mad.nexp"]] <- list(
mapfn = function(resX.mad) { return(exp(-resX.mad)) }
, args = c("resX.mad"))
glbFeatsDerive[["resY.min"]] <- list(
mapfn = function(resYLst) { return(sapply(resYLst, function(thsObsFeat)
min(as.numeric(unlist(str_split(thsObsFeat, ",")))))) }
, args = c("resYLst"))
glbFeatsDerive[["resY.max"]] <- list(
mapfn = function(resYLst) { return(sapply(resYLst, function(thsObsFeat)
max(as.numeric(unlist(str_split(thsObsFeat, ",")))))) }
, args = c("resYLst"))
glbFeatsDerive[["resY.mean"]] <- list(
mapfn = function(resYLst) { return(sapply(resYLst, function(thsObsFeat)
mean(as.numeric(unlist(str_split(thsObsFeat, ",")))))) }
, args = c("resYLst"))
glbFeatsDerive[["resY.mad"]] <- list(
mapfn = function(resYLst) { return(sapply(resYLst, function(thsObsFeat)
mad(as.numeric(unlist(str_split(thsObsFeat, ",")))))) }
, args = c("resYLst"))
glbFeatsDerive[["resY.min.log1p"]] <- list(
mapfn = function(resY.min) { return(log1p(resY.min)) }
, args = c("resY.min"))
glbFeatsDerive[["resY.min.root2"]] <- list(
mapfn = function(resY.min) { return(resY.min ^ (1/2)) }
, args = c("resY.min"))
glbFeatsDerive[["resY.min.nexp"]] <- list(
mapfn = function(resY.min) { return(exp(-resY.min)) }
, args = c("resY.min"))
glbFeatsDerive[["resY.max.log1p"]] <- list(
mapfn = function(resY.max) { return(log1p(resY.max)) }
, args = c("resY.max"))
glbFeatsDerive[["resY.max.root2"]] <- list(
mapfn = function(resY.max) { return(resY.max ^ (1/2)) }
, args = c("resY.max"))
glbFeatsDerive[["resY.max.nexp"]] <- list(
mapfn = function(resY.max) { return(exp(-resY.max)) }
, args = c("resY.max"))
glbFeatsDerive[["resY.mean.log1p"]] <- list(
mapfn = function(resY.mean) { return(log1p(resY.mean)) }
, args = c("resY.mean"))
glbFeatsDerive[["resY.mean.root2"]] <- list(
mapfn = function(resY.mean) { return(resY.mean ^ (1/2)) }
, args = c("resY.mean"))
glbFeatsDerive[["resY.mean.nexp"]] <- list(
mapfn = function(resY.mean) { return(exp(-resY.mean)) }
, args = c("resY.mean"))
glbFeatsDerive[["resY.mad.log1p"]] <- list(
mapfn = function(resY.mad) { return(log1p(resY.mad)) }
, args = c("resY.mad"))
glbFeatsDerive[["resY.mad.root2"]] <- list(
mapfn = function(resY.mad) { return(resY.mad ^ (1/2)) }
, args = c("resY.mad"))
glbFeatsDerive[["resY.mad.nexp"]] <- list(
mapfn = function(resY.mad) { return(exp(-resY.mad)) }
, args = c("resY.mad"))
glbFeatsDerive[["resXY.min"]] <- list(
mapfn = function(resXLst, resYLst) {
resXYAll <- c()
for (obsIx in 1:length(resXLst)) {
resX <- as.numeric(unlist(str_split(resXLst[obsIx], ",")))
resY <- as.numeric(unlist(str_split(resYLst[obsIx], ",")))
resXYAll <- c(resXYAll, min(resX * resY))
}
return(resXYAll)
}
, args = c("resXLst","resYLst"))
glbFeatsDerive[["resXY.max"]] <- list(
mapfn = function(resXLst, resYLst) {
resXYAll <- c()
for (obsIx in 1:length(resXLst)) {
resX <- as.numeric(unlist(str_split(resXLst[obsIx], ",")))
resY <- as.numeric(unlist(str_split(resYLst[obsIx], ",")))
resXYAll <- c(resXYAll, max(resX * resY))
}
return(resXYAll)
}
, args = c("resXLst","resYLst"))
glbFeatsDerive[["resXY.mean"]] <- list(
mapfn = function(resXLst, resYLst) {
resXYAll <- c()
for (obsIx in 1:length(resXLst)) {
resX <- as.numeric(unlist(str_split(resXLst[obsIx], ",")))
resY <- as.numeric(unlist(str_split(resYLst[obsIx], ",")))
resXYAll <- c(resXYAll, mean(resX * resY))
}
return(resXYAll)
}
, args = c("resXLst","resYLst"))
glbFeatsDerive[["resXY.mad"]] <- list(
mapfn = function(resXLst, resYLst) {
resXYAll <- c()
for (obsIx in 1:length(resXLst)) {
resX <- as.numeric(unlist(str_split(resXLst[obsIx], ",")))
resY <- as.numeric(unlist(str_split(resYLst[obsIx], ",")))
resXYAll <- c(resXYAll, mad(resX * resY))
}
return(resXYAll)
}
, args = c("resXLst","resYLst"))
glbFeatsDerive[["resXY.min.log1p"]] <- list(
mapfn = function(resXY.min) { return(log1p(resXY.min)) }
, args = c("resXY.min"))
glbFeatsDerive[["resXY.min.root2"]] <- list(
mapfn = function(resXY.min) { return(resXY.min ^ (1/2)) }
, args = c("resXY.min"))
glbFeatsDerive[["resXY.min.nexp"]] <- list(
mapfn = function(resXY.min) { return(exp(-resXY.min)) }
, args = c("resXY.min"))
glbFeatsDerive[["resXY.max.log1p"]] <- list(
mapfn = function(resXY.max) { return(log1p(resXY.max)) }
, args = c("resXY.max"))
glbFeatsDerive[["resXY.max.root2"]] <- list(
mapfn = function(resXY.max) { return(resXY.max ^ (1/2)) }
, args = c("resXY.max"))
glbFeatsDerive[["resXY.max.nexp"]] <- list(
mapfn = function(resXY.max) { return(exp(-resXY.max)) }
, args = c("resXY.max"))
glbFeatsDerive[["resXY.mean.log1p"]] <- list(
mapfn = function(resXY.mean) { return(log1p(resXY.mean)) }
, args = c("resXY.mean"))
glbFeatsDerive[["resXY.mean.root2"]] <- list(
mapfn = function(resXY.mean) { return(resXY.mean ^ (1/2)) }
, args = c("resXY.mean"))
glbFeatsDerive[["resXY.mean.nexp"]] <- list(
mapfn = function(resXY.mean) { return(exp(-resXY.mean)) }
, args = c("resXY.mean"))
glbFeatsDerive[["resXY.mad.log1p"]] <- list(
mapfn = function(resXY.mad) { return(log1p(resXY.mad)) }
, args = c("resXY.mad"))
glbFeatsDerive[["resXY.mad.root2"]] <- list(
mapfn = function(resXY.mad) { return(resXY.mad ^ (1/2)) }
, args = c("resXY.mad"))
glbFeatsDerive[["resXY.mad.nexp"]] <- list(
mapfn = function(resXY.mad) { return(exp(-resXY.mad)) }
, args = c("resXY.mad"))
# Add logs of numerics that are not distributed normally
# Derive & keep multiple transformations of the same feature, if normality is hard to achieve with just one transformation
# Right skew: logp1; sqrt; ^ 1/3; logp1(logp1); log10; exp(-<feat>/constant)
# glbFeatsDerive[["WordCount.log1p"]] <- list(
# mapfn = function(WordCount) { return(log1p(WordCount)) }
# , args = c("WordCount"))
# glbFeatsDerive[["WordCount.root2"]] <- list(
# mapfn = function(WordCount) { return(WordCount ^ (1/2)) }
# , args = c("WordCount"))
# glbFeatsDerive[["WordCount.nexp"]] <- list(
# mapfn = function(WordCount) { return(exp(-WordCount)) }
# , args = c("WordCount"))
#print(summary(glbObsAll$WordCount))
#print(summary(mapfn(glbObsAll$WordCount)))
# mapfn = function(HOSPI.COST) { return(cut(HOSPI.COST, 5, breaks = c(0, 100000, 200000, 300000, 900000), labels = NULL)) }
# mapfn = function(Rasmussen) { return(ifelse(sign(Rasmussen) >= 0, 1, 0)) }
# mapfn = function(startprice) { return(startprice ^ (1/2)) }
# mapfn = function(startprice) { return(log(startprice)) }
# mapfn = function(startprice) { return(exp(-startprice / 20)) }
# mapfn = function(startprice) { return(scale(log(startprice))) }
# mapfn = function(startprice) { return(sign(sprice.predict.diff) * (abs(sprice.predict.diff) ^ (1/10))) }
# factor
glbFeatsDerive[["lunch"]] <- list(
mapfn = function(labels) { return(factor(
sapply(labels, function(obsLabel) {if (is.na(obsLabel)) return(NA);
ifelse(any(as.numeric(unlist(str_split(obsLabel, " "))) %in% c(0)), "0", "-1") })
, levels = c("-1", "0"))) }
, args = c("labels"))
glbFeatsDerive[["dinner"]] <- list(
mapfn = function(labels) { return(factor(
sapply(labels, function(obsLabel) {if (is.na(obsLabel)) return(NA);
ifelse(any(as.numeric(unlist(str_split(obsLabel, " "))) %in% c(1)), "1", "-1") })
, levels = c("-1", "1"))) }
, args = c("labels"))
glbFeatsDerive[["reserve"]] <- list(
mapfn = function(labels) { return(factor(
sapply(labels, function(obsLabel) {if (is.na(obsLabel)) return(NA);
ifelse(any(as.numeric(unlist(str_split(obsLabel, " "))) %in% c(2)), "2", "-1") })
, levels = c("-1", "2"))) }
, args = c("labels"))
glbFeatsDerive[["outdoor"]] <- list(
mapfn = function(labels) { return(factor(
sapply(labels, function(obsLabel) {if (is.na(obsLabel)) return(NA);
ifelse(any(as.numeric(unlist(str_split(obsLabel, " "))) %in% c(3)), "3", "-1") })
, levels = c("-1", "3"))) }
, args = c("labels"))
glbFeatsDerive[["expensive"]] <- list(
mapfn = function(labels) { return(factor(
sapply(labels, function(obsLabel) {if (is.na(obsLabel)) return(NA);
ifelse(any(as.numeric(unlist(str_split(obsLabel, " "))) %in% c(4)), "4", "-1") })
, levels = c("-1", "4"))) }
, args = c("labels"))
glbFeatsDerive[["liquor"]] <- list(
mapfn = function(labels) { return(factor(
sapply(labels, function(obsLabel) {if (is.na(obsLabel)) return(NA);
ifelse(any(as.numeric(unlist(str_split(obsLabel, " "))) %in% c(5)), "5", "-1") })
, levels = c("-1", "5"))) }
, args = c("labels"))
glbFeatsDerive[["table"]] <- list(
mapfn = function(labels) { return(factor(
sapply(labels, function(obsLabel) {if (is.na(obsLabel)) return(NA);
ifelse(any(as.numeric(unlist(str_split(obsLabel, " "))) %in% c(6)), "6", "-1") })
, levels = c("-1", "6"))) }
, args = c("labels"))
glbFeatsDerive[["classy"]] <- list(
mapfn = function(labels) { return(factor(
sapply(labels, function(obsLabel) {if (is.na(obsLabel)) return(NA);
ifelse(any(as.numeric(unlist(str_split(obsLabel, " "))) %in% c(7)), "7", "-1") })
, levels = c("-1", "7"))) }
, args = c("labels"))
glbFeatsDerive[["kids"]] <- list(
mapfn = function(labels) { return(factor(
sapply(labels, function(obsLabel) {if (is.na(obsLabel)) return(NA);
ifelse(any(as.numeric(unlist(str_split(obsLabel, " "))) %in% c(8)), "8", "-1") })
, levels = c("-1", "8"))) }
, args = c("labels"))
glbFeatsDerive[["nImgs.cut.fctr"]] <- list(
mapfn = function(nImgs) { return(cut(nImgs, breaks = c(0, 32, 60, 120, 3000))) }
, args = c("nImgs"))
# mapfn = function(PropR) { return(as.factor(ifelse(PropR >= 0.5, "Y", "N"))) }
# mapfn = function(productline, description) { as.factor(gsub(" ", "", productline)) }
# mapfn = function(purpose) { return(relevel(as.factor(purpose), ref="all_other")) }
# mapfn = function(raw) { tfr_raw <- as.character(cut(raw, 5));
# tfr_raw[is.na(tfr_raw)] <- "NA.my";
# return(as.factor(tfr_raw)) }
# mapfn = function(startprice.log10) { return(cut(startprice.log10, 3)) }
# mapfn = function(startprice.log10) { return(cut(sprice.predict.diff, c(-1000, -100, -10, -1, 0, 1, 10, 100, 1000))) }
# , args = c("<arg1>"))
# multiple args
# mapfn = function(id, date) { return(paste(as.character(id), as.character(date), sep = "#")) }
# mapfn = function(PTS, oppPTS) { return(PTS - oppPTS) }
# mapfn = function(startprice.log10.predict, startprice) {
# return(spdiff <- (10 ^ startprice.log10.predict) - startprice) }
# mapfn = function(productline, description) { as.factor(
# paste(gsub(" ", "", productline), as.numeric(nchar(description) > 0), sep = "*")) }
# mapfn = function(.src, .pos) {
# return(paste(.src, sprintf("%04d",
# ifelse(.src == "Train", .pos, .pos - 7049)
# ), sep = "#")) }
# # If glbObsAll is not sorted in the desired manner
# mapfn=function(Week) { return(coredata(lag(zoo(orderBy(~Week, glbObsAll)$ILI), -2, na.pad=TRUE))) }
# mapfn=function(ILI) { return(coredata(lag(zoo(ILI), -2, na.pad=TRUE))) }
# mapfn=function(ILI.2.lag) { return(log(ILI.2.lag)) }
# glbFeatsDerive[["<var1>"]] <- glbFeatsDerive[["<var2>"]]
glb_derive_vars <- names(glbFeatsDerive)
# tst <- "descr.my"; args_lst <- NULL; for (arg in glbFeatsDerive[[tst]]$args) args_lst[[arg]] <- glbObsAll[, arg]; print(head(args_lst[[arg]])); print(head(drv_vals <- do.call(glbFeatsDerive[[tst]]$mapfn, args_lst)));
# print(which_ix <- which(args_lst[[arg]] == 0.75)); print(drv_vals[which_ix]);
glbFeatsDateTime <- list()
# glbFeatsDateTime[["<DateTimeFeat>"]] <-
# c(format = "%Y-%m-%d %H:%M:%S", timezone = "America/New_York", impute.na = TRUE,
# last.ctg = TRUE, poly.ctg = TRUE)
glbFeatsPrice <- NULL # or c("<price_var>")
glbFeatsImage <- list() #list(<imageFeat> = list(patchSize = 10)) # if patchSize not specified, no patch computation
glbFeatsText <- list()
Sys.setlocale("LC_ALL", "C") # For english## [1] "C/C/C/C/C/en_US.UTF-8"#glbFeatsText[["<TextFeature>"]] <- list(NULL,
# ,names = myreplacePunctuation(str_to_lower(gsub(" ", "", c(NULL,
# <comma-separated-screened-names>
# ))))
# ,rareWords = myreplacePunctuation(str_to_lower(gsub(" ", "", c(NULL,
# <comma-separated-nonSCOWL-words>
# ))))
#)
# Text Processing Step: custom modifications not present in txt_munge -> use glbFeatsDerive
# Text Processing Step: universal modifications
glb_txt_munge_filenames_pfx <- "<projectId>_mytxt_"
# Text Processing Step: tolower
# Text Processing Step: myreplacePunctuation
# Text Processing Step: removeWords
glb_txt_stop_words <- list()
# Remember to use unstemmed words
if (length(glbFeatsText) > 0) {
require(tm)
require(stringr)
glb_txt_stop_words[["<txt_var>"]] <- sort(myreplacePunctuation(str_to_lower(gsub(" ", "", c(NULL
# Remove any words from stopwords
# , setdiff(myreplacePunctuation(stopwords("english")), c("<keep_wrd1>", <keep_wrd2>"))
# Remove salutations
,"mr","mrs","dr","Rev"
# Remove misc
#,"th" # Happy [[:digit::]]+th birthday
# Remove terms present in Trn only or New only; search for "Partition post-stem"
# ,<comma-separated-terms>
# cor.y.train == NA
# ,unlist(strsplit(paste(c(NULL
# ,"<comma-separated-terms>"
# ), collapse=",")
# freq == 1; keep c("<comma-separated-terms-to-keep>")
# ,<comma-separated-terms>
# chisq.pval high (e.g. == 1); keep c("<comma-separated-terms-to-keep>")
# ,<comma-separated-terms>
# nzv.freqRatio high (e.g. >= glbFeatsNzvFreqMax); keep c("<comma-separated-terms-to-keep>")
# ,<comma-separated-terms>
)))))
}
#orderBy(~term, glb_post_stem_words_terms_df_lst[[txtFeat]][grep("^man", glb_post_stem_words_terms_df_lst[[txtFeat]]$term), ])
#glbObsAll[glb_post_stem_words_terms_mtrx_lst[[txtFeat]][, 4866] > 0, c(glb_rsp_var, txtFeat)]
# To identify terms with a specific freq
#paste0(sort(subset(glb_post_stop_words_terms_df_lst[[txtFeat]], freq == 1)$term), collapse = ",")
#paste0(sort(subset(glb_post_stem_words_terms_df_lst[[txtFeat]], freq <= 2)$term), collapse = ",")
#subset(glb_post_stem_words_terms_df_lst[[txtFeat]], term %in% c("zinger"))
# To identify terms with a specific freq &
# are not stemmed together later OR is value of color.fctr (e.g. gold)
#paste0(sort(subset(glb_post_stop_words_terms_df_lst[[txtFeat]], (freq == 1) & !(term %in% c("blacked","blemish","blocked","blocks","buying","cables","careful","carefully","changed","changing","chargers","cleanly","cleared","connect","connects","connected","contains","cosmetics","default","defaulting","defective","definitely","describe","described","devices","displays","drop","drops","engravement","excellant","excellently","feels","fix","flawlessly","frame","framing","gentle","gold","guarantee","guarantees","handled","handling","having","install","iphone","iphones","keeped","keeps","known","lights","line","lining","liquid","liquidation","looking","lots","manuals","manufacture","minis","most","mostly","network","networks","noted","opening","operated","performance","performs","person","personalized","photograph","physically","placed","places","powering","pre","previously","products","protection","purchasing","returned","rotate","rotation","running","sales","second","seconds","shipped","shuts","sides","skin","skinned","sticker","storing","thats","theres","touching","unusable","update","updates","upgrade","weeks","wrapped","verified","verify") ))$term), collapse = ",")
#print(subset(glb_post_stem_words_terms_df_lst[[txtFeat]], (freq <= 2)))
#glbObsAll[which(terms_mtrx[, 229] > 0), glbFeatsText]
# To identify terms with cor.y == NA
#orderBy(~-freq+term, subset(glb_post_stop_words_terms_df_lst[[txtFeat]], is.na(cor.y)))
#paste(sort(subset(glb_post_stop_words_terms_df_lst[[txtFeat]], is.na(cor.y))[, "term"]), collapse=",")
#orderBy(~-freq+term, subset(glb_post_stem_words_terms_df_lst[[txtFeat]], is.na(cor.y)))
# To identify terms with low cor.y.abs
#head(orderBy(~cor.y.abs+freq+term, subset(glb_post_stem_words_terms_df_lst[[txtFeat]], !is.na(cor.y))), 5)
# To identify terms with high chisq.pval
#subset(glb_post_stem_words_terms_df_lst[[txtFeat]], chisq.pval > 0.99)
#paste0(sort(subset(glb_post_stem_words_terms_df_lst[[txtFeat]], (chisq.pval > 0.99) & (freq <= 10))$term), collapse=",")
#paste0(sort(subset(glb_post_stem_words_terms_df_lst[[txtFeat]], (chisq.pval > 0.9))$term), collapse=",")
#head(orderBy(~-chisq.pval+freq+term, glb_post_stem_words_terms_df_lst[[txtFeat]]), 5)
#glbObsAll[glb_post_stem_words_terms_mtrx_lst[[txtFeat]][, 68] > 0, glbFeatsText]
#orderBy(~term, glb_post_stem_words_terms_df_lst[[txtFeat]][grep("^m", glb_post_stem_words_terms_df_lst[[txtFeat]]$term), ])
# To identify terms with high nzv.freqRatio
#summary(glb_post_stem_words_terms_df_lst[[txtFeat]]$nzv.freqRatio)
#paste0(sort(setdiff(subset(glb_post_stem_words_terms_df_lst[[txtFeat]], (nzv.freqRatio >= glbFeatsNzvFreqMax) & (freq < 10) & (chisq.pval >= 0.05))$term, c( "128gb","3g","4g","gold","ipad1","ipad3","ipad4","ipadair2","ipadmini2","manufactur","spacegray","sprint","tmobil","verizon","wifion"))), collapse=",")
# To identify obs with a txt term
#tail(orderBy(~-freq+term, glb_post_stop_words_terms_df_lst[[txtFeat]]), 20)
#mydspObs(list(descr.my.contains="non"), cols=c("color", "carrier", "cellular", "storage"))
#grep("ever", dimnames(terms_stop_mtrx)$Terms)
#which(terms_stop_mtrx[, grep("ipad", dimnames(terms_stop_mtrx)$Terms)] > 0)
#glbObsAll[which(terms_stop_mtrx[, grep("16", dimnames(terms_stop_mtrx)$Terms)[1]] > 0), c(glbFeatsCategory, "storage", txtFeat)]
# Text Processing Step: screen for names # Move to glbFeatsText specs section in order of text processing steps
# glbFeatsText[["<txtFeat>"]]$names <- myreplacePunctuation(str_to_lower(gsub(" ", "", c(NULL
# # Person names for names screening
# ,<comma-separated-list>
#
# # Company names
# ,<comma-separated-list>
#
# # Product names
# ,<comma-separated-list>
# ))))
# glbFeatsText[["<txtFeat>"]]$rareWords <- myreplacePunctuation(str_to_lower(gsub(" ", "", c(NULL
# # Words not in SCOWL db
# ,<comma-separated-list>
# ))))
# To identify char vectors post glbFeatsTextMap
#grep("six(.*)hour", glb_txt_chr_lst[[txtFeat]], ignore.case = TRUE, value = TRUE)
#grep("[S|s]ix(.*)[H|h]our", glb_txt_chr_lst[[txtFeat]], value = TRUE)
# To identify whether terms shd be synonyms
#orderBy(~term, glb_post_stop_words_terms_df_lst[[txtFeat]][grep("^moder", glb_post_stop_words_terms_df_lst[[txtFeat]]$term), ])
# term_row_df <- glb_post_stop_words_terms_df_lst[[txtFeat]][grep("^came$", glb_post_stop_words_terms_df_lst[[txtFeat]]$term), ]
#
# cor(glb_post_stop_words_terms_mtrx_lst[[txtFeat]][glbObsAll$.lcn == "Fit", term_row_df$pos], glbObsTrn[, glb_rsp_var], use="pairwise.complete.obs")
# To identify which stopped words are "close" to a txt term
#sort(cluster_vars)
# Text Processing Step: stemDocument
# To identify stemmed txt terms
#glb_post_stop_words_terms_df_lst[[txtFeat]][grep("^la$", glb_post_stop_words_terms_df_lst[[txtFeat]]$term), ]
#orderBy(~term, glb_post_stem_words_terms_df_lst[[txtFeat]][grep("^con", glb_post_stem_words_terms_df_lst[[txtFeat]]$term), ])
#glbObsAll[which(terms_stem_mtrx[, grep("use", dimnames(terms_stem_mtrx)$Terms)[[1]]] > 0), c(glbFeatsId, "productline", txtFeat)]
#glbObsAll[which(TfIdf_stem_mtrx[, 191] > 0), c(glbFeatsId, glbFeatsCategory, txtFeat)]
#glbObsAll[which(glb_post_stop_words_terms_mtrx_lst[[txtFeat]][, 6165] > 0), c(glbFeatsId, glbFeatsCategory, txtFeat)]
#which(glbObsAll$UniqueID %in% c(11915, 11926, 12198))
# Text Processing Step: mycombineSynonyms
# To identify which terms are associated with not -> combine "could not" & "couldn't"
#findAssocs(glb_full_DTM_lst[[txtFeat]], "not", 0.05)
# To identify which synonyms should be combined
#orderBy(~term, glb_post_stem_words_terms_df_lst[[txtFeat]][grep("^c", glb_post_stem_words_terms_df_lst[[txtFeat]]$term), ])
chk_comb_cor <- function(syn_lst) {
# cor(terms_stem_mtrx[glbObsAll$.src == "Train", grep("^(damag|dent|ding)$", dimnames(terms_stem_mtrx)[[2]])], glbObsTrn[, glb_rsp_var], use="pairwise.complete.obs")
print(subset(glb_post_stem_words_terms_df_lst[[txtFeat]], term %in% syn_lst$syns))
print(subset(get_corpus_terms(tm_map(glbFeatsTextCorpus[[txtFeat]], mycombineSynonyms, list(syn_lst), lazy=FALSE)), term == syn_lst$word))
# cor(terms_stop_mtrx[glbObsAll$.src == "Train", grep("^(damage|dent|ding)$", dimnames(terms_stop_mtrx)[[2]])], glbObsTrn[, glb_rsp_var], use="pairwise.complete.obs")
# cor(rowSums(terms_stop_mtrx[glbObsAll$.src == "Train", grep("^(damage|dent|ding)$", dimnames(terms_stop_mtrx)[[2]])]), glbObsTrn[, glb_rsp_var], use="pairwise.complete.obs")
}
#chk_comb_cor(syn_lst=list(word="cabl", syns=c("cabl", "cord")))
#chk_comb_cor(syn_lst=list(word="damag", syns=c("damag", "dent", "ding")))
#chk_comb_cor(syn_lst=list(word="dent", syns=c("dent", "ding")))
#chk_comb_cor(syn_lst=list(word="use", syns=c("use", "usag")))
glbFeatsTextSynonyms <- list()
# list parsed to collect glbFeatsText[[<txtFeat>]]$vldTerms
# glbFeatsTextSynonyms[["Hdln.my"]] <- list(NULL
# # people in places
# , list(word = "australia", syns = c("australia", "australian"))
# , list(word = "italy", syns = c("italy", "Italian"))
# , list(word = "newyork", syns = c("newyork", "newyorker"))
# , list(word = "Pakistan", syns = c("Pakistan", "Pakistani"))
# , list(word = "peru", syns = c("peru", "peruvian"))
# , list(word = "qatar", syns = c("qatar", "qatari"))
# , list(word = "scotland", syns = c("scotland", "scotish"))
# , list(word = "Shanghai", syns = c("Shanghai", "Shanzhai"))
# , list(word = "venezuela", syns = c("venezuela", "venezuelan"))
#
# # companies - needs to be data dependent
# # - e.g. ensure BNP in this experiment/feat always refers to BNPParibas
#
# # general synonyms
# , list(word = "Create", syns = c("Create","Creator"))
# , list(word = "cute", syns = c("cute","cutest"))
# , list(word = "Disappear", syns = c("Disappear","Fadeout"))
# , list(word = "teach", syns = c("teach", "taught"))
# , list(word = "theater", syns = c("theater", "theatre", "theatres"))
# , list(word = "understand", syns = c("understand", "understood"))
# , list(word = "weak", syns = c("weak", "weaken", "weaker", "weakest"))
# , list(word = "wealth", syns = c("wealth", "wealthi"))
#
# # custom synonyms (phrases)
#
# # custom synonyms (names)
# )
#glbFeatsTextSynonyms[["<txtFeat>"]] <- list(NULL
# , list(word="<stem1>", syns=c("<stem1>", "<stem1_2>"))
# )
for (txtFeat in names(glbFeatsTextSynonyms))
for (entryIx in 1:length(glbFeatsTextSynonyms[[txtFeat]])) {
glbFeatsTextSynonyms[[txtFeat]][[entryIx]]$word <-
str_to_lower(glbFeatsTextSynonyms[[txtFeat]][[entryIx]]$word)
glbFeatsTextSynonyms[[txtFeat]][[entryIx]]$syns <-
str_to_lower(glbFeatsTextSynonyms[[txtFeat]][[entryIx]]$syns)
}
glbFeatsTextSeed <- 181
# tm options include: check tm::weightSMART
glb_txt_terms_control <- list( # Gather model performance & run-time stats
# weighting = function(x) weightSMART(x, spec = "nnn")
# weighting = function(x) weightSMART(x, spec = "lnn")
# weighting = function(x) weightSMART(x, spec = "ann")
# weighting = function(x) weightSMART(x, spec = "bnn")
# weighting = function(x) weightSMART(x, spec = "Lnn")
#
weighting = function(x) weightSMART(x, spec = "ltn") # default
# weighting = function(x) weightSMART(x, spec = "lpn")
#
# weighting = function(x) weightSMART(x, spec = "ltc")
#
# weighting = weightBin
# weighting = weightTf
# weighting = weightTfIdf # : default
# termFreq selection criteria across obs: tm default: list(global=c(1, Inf))
, bounds = list(global = c(1, Inf))
# wordLengths selection criteria: tm default: c(3, Inf)
, wordLengths = c(1, Inf)
)
glb_txt_cor_var <- glb_rsp_var # : default # or c(<feat>)
# select one from c("union.top.val.cor", "top.cor", "top.val", default: "top.chisq", "sparse")
glbFeatsTextFilter <- "top.chisq"
glbFeatsTextTermsMax <- rep(10, length(glbFeatsText)) # :default
names(glbFeatsTextTermsMax) <- names(glbFeatsText)
# Text Processing Step: extractAssoc
glbFeatsTextAssocCor <- rep(1, length(glbFeatsText)) # :default
names(glbFeatsTextAssocCor) <- names(glbFeatsText)
# Remember to use stemmed terms
glb_important_terms <- list()
# Text Processing Step: extractPatterns (ngrams)
glbFeatsTextPatterns <- list()
#glbFeatsTextPatterns[[<txtFeat>>]] <- list()
#glbFeatsTextPatterns[[<txtFeat>>]] <- c(metropolitan.diary.colon = "Metropolitan Diary:")
# Have to set it even if it is not used
# Properties:
# numrows(glb_feats_df) << numrows(glbObsFit
# Select terms that appear in at least 0.2 * O(FP/FN(glbObsOOB)) ???
# numrows(glbObsOOB) = 1.1 * numrows(glbObsNew) ???
glb_sprs_thresholds <- NULL # or c(<txtFeat1> = 0.988, <txtFeat2> = 0.970, <txtFeat3> = 0.970)
glbFctrMaxUniqVals <- 20 # default: 20
glb_impute_na_data <- FALSE # or TRUE
glb_mice_complete.seed <- 144 # or any integer
glb_cluster <- FALSE # : default or TRUE
glb_cluster.seed <- 189 # or any integer
glb_cluster_entropy_var <- NULL # c(glb_rsp_var, as.factor(cut(glb_rsp_var, 3)), default: NULL)
glbFeatsTextClusterVarsExclude <- FALSE # default FALSE
glb_interaction_only_feats <- NULL # : default or c(<parent_feat> = "<child_feat>")
glbFeatsNzvFreqMax <- 19 # 19 : caret default
glbFeatsNzvUniqMin <- 10 # 10 : caret default
glbRFESizes <- list()
#glbRFESizes[["mdlFamily"]] <- c(4, 8, 16, 32, 64, 67, 68, 69) # Accuracy@69/70 = 0.8258
glbObsFitOutliers <- list()
# If outliers.n >= 10; consider concatenation of interaction vars
# glbObsFitOutliers[["<mdlFamily>"]] <- c(NULL
# is.na(.rstudent)
# max(.rstudent)
# is.na(.dffits)
# .hatvalues >= 0.99
# -38,167,642 < minmax(.rstudent) < 49,649,823
# , <comma-separated-<glbFeatsId>>
# )
glbObsTrnOutliers <- list()
glbObsTrnOutliers[["Final"]] <- union(glbObsFitOutliers[["All.X"]],
c(NULL
))
# influence.measures: car::outlier; rstudent; dffits; hatvalues; dfbeta; dfbetas
#mdlId <- "All.X##rcv#glm"; obs_df <- fitobs_df
#mdlId <- "RFE.X.glm"; obs_df <- fitobs_df
#mdlId <- "Final.glm"; obs_df <- trnobs_df
#mdlId <- "CSM2.X.glm"; obs_df <- fitobs_df
#print(outliers <- car::outlierTest(glb_models_lst[[mdlId]]$finalModel))
#mdlIdFamily <- paste0(head(unlist(str_split(mdlId, "\\.")), -1), collapse="."); obs_df <- dplyr::filter_(obs_df, interp(~(!(var %in% glbObsFitOutliers[[mdlIdFamily]])), var = as.name(glbFeatsId))); model_diags_df <- cbind(obs_df, data.frame(.rstudent=stats::rstudent(glb_models_lst[[mdlId]]$finalModel)), data.frame(.dffits=stats::dffits(glb_models_lst[[mdlId]]$finalModel)), data.frame(.hatvalues=stats::hatvalues(glb_models_lst[[mdlId]]$finalModel)));print(summary(model_diags_df[, c(".rstudent",".dffits",".hatvalues")])); table(cut(model_diags_df$.hatvalues, breaks=c(0.00, 0.98, 0.99, 1.00)))
#print(subset(model_diags_df, is.na(.rstudent))[, glbFeatsId])
#print(model_diags_df[which.max(model_diags_df$.rstudent), ])
#print(subset(model_diags_df, is.na(.dffits))[, glbFeatsId])
#print(model_diags_df[which.min(model_diags_df$.dffits), ])
#print(subset(model_diags_df, .hatvalues > 0.99)[, glbFeatsId])
#dffits_df <- merge(dffits_df, outliers_df, by="row.names", all.x=TRUE); row.names(dffits_df) <- dffits_df$Row.names; dffits_df <- subset(dffits_df, select=-Row.names)
#dffits_df <- merge(dffits_df, glbObsFit, by="row.names", all.x=TRUE); row.names(dffits_df) <- dffits_df$Row.names; dffits_df <- subset(dffits_df, select=-Row.names)
#subset(dffits_df, !is.na(.Bonf.p))
#mdlId <- "CSM.X.glm"; vars <- myextract_actual_feats(row.names(orderBy(reformulate(c("-", paste0(mdlId, ".imp"))), myget_feats_imp(glb_models_lst[[mdlId]]))));
#model_diags_df <- glb_get_predictions(model_diags_df, mdlId, glb_rsp_var)
#obs_ix <- row.names(model_diags_df) %in% names(outliers$rstudent)[1]
#obs_ix <- which(is.na(model_diags_df$.rstudent))
#obs_ix <- which(is.na(model_diags_df$.dffits))
#myplot_parcoord(obs_df=model_diags_df[, c(glbFeatsId, glbFeatsCategory, ".rstudent", ".dffits", ".hatvalues", glb_rsp_var, paste0(glb_rsp_var, mdlId), vars[1:min(20, length(vars))])], obs_ix=obs_ix, id_var=glbFeatsId, category_var=glbFeatsCategory)
#model_diags_df[row.names(model_diags_df) %in% names(outliers$rstudent)[c(1:2)], ]
#ctgry_diags_df <- model_diags_df[model_diags_df[, glbFeatsCategory] %in% c("Unknown#0"), ]
#myplot_parcoord(obs_df=ctgry_diags_df[, c(glbFeatsId, glbFeatsCategory, ".rstudent", ".dffits", ".hatvalues", glb_rsp_var, "startprice.log10.predict.RFE.X.glmnet", indep_vars[1:20])], obs_ix=row.names(ctgry_diags_df) %in% names(outliers$rstudent)[1], id_var=glbFeatsId, category_var=glbFeatsCategory)
#table(glbObsFit[model_diags_df[, glbFeatsCategory] %in% c("iPad1#1"), "startprice.log10.cut.fctr"])
#glbObsFit[model_diags_df[, glbFeatsCategory] %in% c("iPad1#1"), c(glbFeatsId, "startprice")]
# No outliers & .dffits == NaN
#myplot_parcoord(obs_df=model_diags_df[, c(glbFeatsId, glbFeatsCategory, glb_rsp_var, "startprice.log10.predict.RFE.X.glmnet", indep_vars[1:10])], obs_ix=seq(1:nrow(model_diags_df))[is.na(model_diags_df$.dffits)], id_var=glbFeatsId, category_var=glbFeatsCategory)
# Modify mdlId to (build & extract) "<FamilyId>#<Fit|Trn>#<caretMethod>#<preProc1.preProc2>#<samplingMethod>"
glb_models_lst <- list(); glb_models_df <- data.frame()
# Regression
if (glb_is_regression) {
glbMdlMethods <- c(NULL
# deterministic
#, "lm", # same as glm
, "glm", "bayesglm", "glmnet"
, "rpart"
# non-deterministic
, "gbm", "rf"
# Unknown
, "nnet" , "avNNet" # runs 25 models per cv sample for tunelength=5
, "svmLinear", "svmLinear2"
, "svmPoly" # runs 75 models per cv sample for tunelength=5
, "svmRadial"
, "earth"
, "bagEarth" # Takes a long time
)
} else
# Classification - Add ada (auto feature selection)
if (glb_is_binomial)
glbMdlMethods <- c(NULL
# deterministic
, "bagEarth" # Takes a long time
, "glm", "bayesglm", "glmnet"
, "nnet"
, "rpart"
# non-deterministic
, "gbm"
, "avNNet" # runs 25 models per cv sample for tunelength=5
, "rf"
# Unknown
, "lda", "lda2"
# svm models crash when predict is called -> internal to kernlab it should call predict without .outcome
, "svmLinear", "svmLinear2"
, "svmPoly" # runs 75 models per cv sample for tunelength=5
, "svmRadial"
, "earth"
) else
glbMdlMethods <- c(NULL
# deterministic
,"glmnet"
# non-deterministic
,"rf"
# Unknown
,"gbm","rpart"
)
glbMdlFamilies <- list(); glb_mdl_feats_lst <- list()
# family: Choose from c("RFE.X", "CSM.X", "All.X", "Best.Interact")
# methods: Choose from c(NULL, <method>, glbMdlMethods)
#glbMdlFamilies[["RFE.X"]] <- c("glmnet", "glm") # non-NULL vector is mandatory
glbMdlFamilies[["All.X"]] <- c("glmnet", "glm") # non-NULL vector is mandatory
#glbMdlFamilies[["Best.Interact"]] <- "glmnet" # non-NULL vector is mandatory
# Check if interaction features make RFE better
# glbMdlFamilies[["CSM.X"]] <- setdiff(glbMdlMethods, c("lda", "lda2")) # crashing due to category:.clusterid ??? #c("glmnet", "glm") # non-NULL list is mandatory
# glb_mdl_feats_lst[["CSM.X"]] <- c(NULL
# , <comma-separated-features-vector>
# )
# dAFeats.CSM.X %<d-% c(NULL
# # Interaction feats up to varImp(RFE.X.glmnet) >= 50
# , <comma-separated-features-vector>
# , setdiff(myextract_actual_feats(predictors(rfe_fit_results)), c(NULL
# , <comma-separated-features-vector>
# ))
# )
# glb_mdl_feats_lst[["CSM.X"]] <- "%<d-% dAFeats.CSM.X"
glbMdlFamilies[["Final"]] <- c(NULL) # NULL vector acceptable # c("glmnet", "glm")
glbMdlAllowParallel <- list()
#glbMdlAllowParallel[["<mdlId>"]] <- FALSE
# Check if tuning parameters make fit better; make it mdlFamily customizable ?
glbMdlTuneParams <- data.frame()
# When glmnet crashes at model$grid with error: ???
glmnetTuneParams <- rbind(data.frame()
,data.frame(parameter = "alpha", vals = "0.100 0.325 0.550 0.775 1.000")
,data.frame(parameter = "lambda", vals = "9.342e-02")
)
# glbMdlTuneParams <- myrbind_df(glbMdlTuneParams,
# cbind(data.frame(mdlId = "<mdlId>"),
# glmnetTuneParams))
#avNNet
# size=[1] 3 5 7 9; decay=[0] 1e-04 0.001 0.01 0.1; bag=[FALSE]; RMSE=1.3300906
#bagEarth
# degree=1 [2] 3; nprune=64 128 256 512 [1024]; RMSE=0.6486663 (up)
# glbMdlTuneParams <- myrbind_df(glbMdlTuneParams, rbind(data.frame()
# ,data.frame(method = "bagEarth", parameter = "nprune", vals = "256")
# ,data.frame(method = "bagEarth", parameter = "degree", vals = "2")
# ))
#earth
# degree=[1]; nprune=2 [9] 17 25 33; RMSE=0.1334478
#gbm
# shrinkage=0.05 [0.10] 0.15 0.20 0.25; n.trees=100 150 200 [250] 300; interaction.depth=[1] 2 3 4 5; n.minobsinnode=[10]; RMSE=0.2008313
# glbMdlTuneParams <- myrbind_df(glbMdlTuneParams, rbind(data.frame()
# ,data.frame(method = "gbm", parameter = "shrinkage", min = 0.05, max = 0.25, by = 0.05)
# ,data.frame(method = "gbm", parameter = "n.trees", min = 100, max = 300, by = 50)
# ,data.frame(method = "gbm", parameter = "interaction.depth", min = 1, max = 5, by = 1)
# ,data.frame(method = "gbm", parameter = "n.minobsinnode", min = 10, max = 10, by = 10)
# #seq(from=0.05, to=0.25, by=0.05)
# ))
#glmnet
# alpha=0.100 [0.325] 0.550 0.775 1.000; lambda=0.0005232693 0.0024288010 0.0112734954 [0.0523269304] 0.2428800957; RMSE=0.6164891
# glbMdlTuneParams <- myrbind_df(glbMdlTuneParams, rbind(data.frame()
# ,data.frame(method = "glmnet", parameter = "alpha", vals = "0.550 0.775 0.8875 0.94375 1.000")
# ,data.frame(method = "glmnet", parameter = "lambda", vals = "9.858855e-05 0.0001971771 0.0009152152 0.0042480525 0.0197177130")
# ))
#nnet
# size=3 5 [7] 9 11; decay=0.0001 0.001 0.01 [0.1] 0.2; RMSE=0.9287422
# glbMdlTuneParams <- myrbind_df(glbMdlTuneParams, rbind(data.frame()
# ,data.frame(method = "nnet", parameter = "size", vals = "3 5 7 9 11")
# ,data.frame(method = "nnet", parameter = "decay", vals = "0.0001 0.0010 0.0100 0.1000 0.2000")
# ))
#rf # Don't bother; results are not deterministic
# mtry=2 35 68 [101] 134; RMSE=0.1339974
# glbMdlTuneParams <- myrbind_df(glbMdlTuneParams, rbind(data.frame()
# ,data.frame(method = "rf", parameter = "mtry", vals = "2 5 9 13 17")
# ))
#rpart
# cp=0.020 [0.025] 0.030 0.035 0.040; RMSE=0.1770237
# glbMdlTuneParams <- myrbind_df(glbMdlTuneParams, rbind(data.frame()
# ,data.frame(method = "rpart", parameter = "cp", vals = "0.004347826 0.008695652 0.017391304 0.021739130 0.034782609")
# ))
#svmLinear
# C=0.01 0.05 [0.10] 0.50 1.00 2.00 3.00 4.00; RMSE=0.1271318; 0.1296718
# glbMdlTuneParams <- myrbind_df(glbMdlTuneParams, rbind(data.frame()
# ,data.frame(method = "svmLinear", parameter = "C", vals = "0.01 0.05 0.1 0.5 1")
# ))
#svmLinear2
# cost=0.0625 0.1250 [0.25] 0.50 1.00; RMSE=0.1276354
# glbMdlTuneParams <- myrbind_df(glbMdlTuneParams, rbind(data.frame()
# ,data.frame(method = "svmLinear2", parameter = "cost", vals = "0.0625 0.125 0.25 0.5 1")
# ))
#svmPoly
# degree=[1] 2 3 4 5; scale=0.01 0.05 [0.1] 0.5 1; C=0.50 1.00 [2.00] 3.00 4.00; RMSE=0.1276130
# glbMdlTuneParams <- myrbind_df(glbMdlTuneParams, rbind(data.frame()
# ,data.frame(method="svmPoly", parameter="degree", min=1, max=5, by=1) #seq(1, 5, 1)
# ,data.frame(method="svmPoly", parameter="scale", vals="0.01, 0.05, 0.1, 0.5, 1")
# ,data.frame(method="svmPoly", parameter="C", vals="0.50, 1.00, 2.00, 3.00, 4.00")
# ))
#svmRadial
# sigma=[0.08674323]; C=0.25 0.50 1.00 [2.00] 4.00; RMSE=0.1614957
#glb2Sav(); all.equal(sav_models_df, glb_models_df)
glb_preproc_methods <- NULL
# c("YeoJohnson", "center.scale", "range", "pca", "ica", "spatialSign")
# Baseline prediction model feature(s)
glb_Baseline_mdl_var <- NULL # or c("<feat>")
glbMdlMetric_terms <- NULL # or matrix(c(
# 0,1,2,3,4,
# 2,0,1,2,3,
# 4,2,0,1,2,
# 6,4,2,0,1,
# 8,6,4,2,0
# ), byrow=TRUE, nrow=5)
glbMdlMetricSummary <- NULL # or "<metric_name>"
glbMdlMetricMaximize <- NULL # or FALSE (TRUE is not the default for both classification & regression)
glbMdlMetricSummaryFn <- NULL # or function(data, lev=NULL, model=NULL) {
# confusion_mtrx <- t(as.matrix(confusionMatrix(data$pred, data$obs)))
# #print(confusion_mtrx)
# #print(confusion_mtrx * glbMdlMetric_terms)
# metric <- sum(confusion_mtrx * glbMdlMetric_terms) / nrow(data)
# names(metric) <- glbMdlMetricSummary
# return(metric)
# }
glbMdlCheckRcv <- FALSE # Turn it on when needed; otherwise takes long time
glb_rcv_n_folds <- 3 # or NULL
glb_rcv_n_repeats <- 3 # or NULL
glb_clf_proba_threshold <- NULL # 0.5
# Model selection criteria
if (glb_is_regression)
glbMdlMetricsEval <- c("min.RMSE.OOB", "max.R.sq.OOB", "max.Adj.R.sq.fit", "min.RMSE.fit")
#glbMdlMetricsEval <- c("min.RMSE.fit", "max.R.sq.fit", "max.Adj.R.sq.fit")
if (glb_is_classification) {
if (glb_is_binomial)
glbMdlMetricsEval <-
c("max.Accuracy.OOB", "max.AUCROCR.OOB", "max.AUCpROC.OOB", "min.aic.fit", "max.Accuracy.fit") else
glbMdlMetricsEval <- c("max.Accuracy.OOB", "max.Kappa.OOB")
}
# select from NULL [no ensemble models], "auto" [all models better than MFO or Baseline], c(mdl_ids in glb_models_lst) [Typically top-rated models in auto]
glb_mdl_ensemble <- NULL
# "%<d-% setdiff(mygetEnsembleAutoMdlIds(), 'CSM.X.rf')"
# c(<comma-separated-mdlIds>
# )
# Only for classifications; for regressions remove "(.*)\\.prob" form the regex
# tmp_fitobs_df <- glbObsFit[, grep(paste0("^", gsub(".", "\\.", mygetPredictIds$value, fixed = TRUE), "CSM\\.X\\.(.*)\\.prob"), names(glbObsFit), value = TRUE)]; cor_mtrx <- cor(tmp_fitobs_df); cor_vctr <- sort(cor_mtrx[row.names(orderBy(~-Overall, varImp(glb_models_lst[["Ensemble.repeatedcv.glmnet"]])$imp))[1], ]); summary(cor_vctr); cor_vctr
#ntv.glm <- glm(reformulate(indep_vars, glb_rsp_var), family = "binomial", data = glbObsFit)
#step.glm <- step(ntv.glm)
glb_sel_mdl_id <- "All.X##rcv#glmnet" #select from c(NULL, "All.X##rcv#glmnet", "RFE.X##rcv#glmnet", <mdlId>)
glb_fin_mdl_id <- NULL #select from c(NULL, glb_sel_mdl_id)
glb_dsp_cols <- c(".pos", glbFeatsId, glbFeatsCategory, glb_rsp_var
# List critical cols excl. above
)
# Output specs
# lclgetfltout_df <- function(obsout_df) {
# require(tidyr)
# obsout_df <- obsout_df %>%
# tidyr::separate("ImageId.x.y", c(".src", ".pos", "x", "y"),
# sep = "#", remove = TRUE, extra = "merge")
#
# # dplyr::summarize(xMeanN = mean(as.numeric(x)), yMeanN = mean(as.numeric(y)))
#
# return(fmnout_df)
# }
glbObsOut <- list(NULL
# glbFeatsId will be the first output column, by default
,vars = list()
# ,mapFn = function(obsout_df) {
# }
)
#obsout_df <- savobsout_df
glbObsOut$mapFn <- function(obsout_df) {
set.seed(997)
txfout_df <- obsout_df %>%
dplyr::mutate(
lunch = levels(glbObsTrn[, "lunch" ])[
round(mean(as.numeric(glbObsTrn[, "lunch" ])), 0)],
dinner = levels(glbObsTrn[, "dinner" ])[
round(mean(as.numeric(glbObsTrn[, "dinner" ])), 0)],
reserve = levels(glbObsTrn[, "reserve" ])[
round(mean(as.numeric(glbObsTrn[, "reserve" ])), 0)],
# outdoor = levels(glbObsTrn[, "outdoor" ])[
# rbinom(nrow(obsout_df), 1, mean(as.numeric(glbObsTrn[, "outdoor" ])) - 1) + 1],
outdoor =
ifelse(levels(glbObsTrn[, "outdoor.fctr" ])[as.numeric(outdoor.fctr)] == "N", "-1", "3"),
expensive = levels(glbObsTrn[, "expensive"])[
round(mean(as.numeric(glbObsTrn[, "expensive"])), 0)],
liquor = levels(glbObsTrn[, "liquor" ])[
round(mean(as.numeric(glbObsTrn[, "liquor" ])), 0)],
table = levels(glbObsTrn[, "table" ])[
round(mean(as.numeric(glbObsTrn[, "table" ])), 0)],
classy = levels(glbObsTrn[, "classy" ])[
round(mean(as.numeric(glbObsTrn[, "classy" ])), 0)],
kids = levels(glbObsTrn[, "kids" ])[
round(mean(as.numeric(glbObsTrn[, "kids" ])), 0)]
)
print("ObsNew output class tables:")
print(sapply(c("lunch","dinner","reserve","outdoor",
"expensive","liquor","table",
"classy","kids"),
function(feat) table(txfout_df[, feat], useNA = "ifany")))
txfout_df <- txfout_df %>%
dplyr::mutate(labels = "") %>%
dplyr::mutate(labels =
ifelse(lunch != "-1", paste(labels, lunch ), labels)) %>%
dplyr::mutate(labels =
ifelse(dinner != "-1", paste(labels, dinner ), labels)) %>%
dplyr::mutate(labels =
ifelse(reserve != "-1", paste(labels, reserve ), labels)) %>%
dplyr::mutate(labels =
ifelse(outdoor != "-1", paste(labels, outdoor ), labels)) %>%
dplyr::mutate(labels =
ifelse(expensive != "-1", paste(labels, expensive), labels)) %>%
dplyr::mutate(labels =
ifelse(liquor != "-1", paste(labels, liquor ), labels)) %>%
dplyr::mutate(labels =
ifelse(table != "-1", paste(labels, table ), labels)) %>%
dplyr::mutate(labels =
ifelse(classy != "-1", paste(labels, classy ), labels)) %>%
dplyr::mutate(labels =
ifelse(kids != "-1", paste(labels, kids ), labels)) %>%
dplyr::select(business_id, labels)
return(txfout_df)
}
#if (!is.null(glbObsOut$mapFn)) obsout_df <- glbObsOut$mapFn(obsout_df); print(head(obsout_df))
glb_out_obs <- NULL # select from c(NULL : default to "new", "all", "new", "trn")
if (glb_is_classification && glb_is_binomial) {
# glbObsOut$vars[["Proba.Y"]] <-
# "%<d-% glbObsNew[, mygetPredictIds(glb_rsp_var, glb_fin_mdl_id)$prob]"
glbObsOut$vars[[glb_rsp_var]] <-
"%<d-% glbObsNew[, mygetPredictIds(glb_rsp_var, glb_fin_mdl_id)$value]"
} else {
# glbObsOut$vars[[glbFeatsId]] <-
# "%<d-% as.integer(gsub('Test#', '', glbObsNew[, glbFeatsId]))"
glbObsOut$vars[[glb_rsp_var]] <-
"%<d-% glbObsNew[, mygetPredictIds(glb_rsp_var, glb_fin_mdl_id)$value]"
# for (outVar in setdiff(glbFeatsExcludeLcl, glb_rsp_var_raw))
# glbObsOut$vars[[outVar]] <-
# paste0("%<d-% mean(glbObsAll[, \"", outVar, "\"], na.rm = TRUE)")
}
# glbObsOut$vars[[glb_rsp_var_raw]] <- glb_rsp_var_raw
# glbObsOut$vars[[paste0(head(unlist(strsplit(mygetPredictIds$value, "")), -1), collapse = "")]] <-
glbOutStackFnames <- NULL #: default
# c("ebayipads_txt_assoc1_out_bid1_stack.csv") # manual stack
# c("ebayipads_finmdl_bid1_out_nnet_1.csv") # universal stack
glbOut <- list(pfx = "YelpRest_resXY_outdoor_")
# lclImageSampleSeed <- 129
glbOutDataVizFname <- NULL # choose from c(NULL, "<projectId>_obsall.csv")
glbChunks <- list(labels = c("set_global_options_wd","set_global_options"
,"import.data","inspect.data","scrub.data","transform.data"
,"extract.features"
,"extract.features.datetime","extract.features.image","extract.features.price"
,"extract.features.text","extract.features.string"
,"extract.features.end"
,"manage.missing.data","cluster.data","partition.data.training","select.features"
,"fit.models_0","fit.models_1","fit.models_2","fit.models_3"
,"fit.data.training_0","fit.data.training_1"
,"predict.data.new"
,"display.session.info"))
# To ensure that all chunks in this script are in glbChunks
if (!is.null(chkChunksLabels <- knitr::all_labels()) && # knitr::all_labels() doesn't work in console runs
!identical(chkChunksLabels, glbChunks$labels)) {
print(sprintf("setdiff(chkChunksLabels, glbChunks$labels): %s",
setdiff(chkChunksLabels, glbChunks$labels)))
print(sprintf("setdiff(glbChunks$labels, chkChunksLabels): %s",
setdiff(glbChunks$labels, chkChunksLabels)))
}
glbChunks[["first"]] <- NULL #default: script will load envir from previous chunk
glbChunks[["last"]] <- NULL #"extract.features.end" #NULL #default: script will save envir at end of this chunk
#mysavChunk(glbOut$pfx, glbChunks[["last"]])
# Inspect max OOB FP
#chkObsOOB <- subset(glbObsOOB, !label.fctr.All.X..rcv.glmnet.is.acc)
#chkObsOOBFP <- subset(chkObsOOB, label.fctr.All.X..rcv.glmnet == "left_eye_center") %>% dplyr::mutate(Probability1 = label.fctr.All.X..rcv.glmnet.prob) %>% select(-.src, -.pos, -x, -y) %>% lclgetfltout_df() %>% mutate(obj.distance = (((as.numeric(x) - left_eye_center_x.int) ^ 2) + ((as.numeric(y) - left_eye_center_y.int) ^ 2)) ^ 0.5) %>% dplyr::top_n(5, obj.distance) %>% dplyr::top_n(5, -patch.cor)
#
#newImgObs <- glbObsNew[(glbObsNew$ImageId == "Test#0001"), ]; print(newImgObs[which.max(newImgObs$label.fctr.Final..rcv.glmnet.prob), ])
#OOBImgObs <- glbObsOOB[(glbObsOOB$ImageId == "Train#0003"), ]; print(OOBImgObs[which.max(OOBImgObs$label.fctr.All.X..rcv.glmnet.prob), ])
#load("<scriptName>_extract.features.end.RData", verbose = TRUE)
#mygetImage(which(glbObsAll[, glbFeatsId] == "Train#0003"), names(glbFeatsImage)[1], plot = TRUE, featHighlight = c("left_eye_center_x", "left_eye_center_y"), ovrlHighlight = c(66, 35))
# Depict process
glb_analytics_pn <- petrinet(name = "glb_analytics_pn",
trans_df = data.frame(id = 1:6,
name = c("data.training.all","data.new",
"model.selected","model.final",
"data.training.all.prediction","data.new.prediction"),
x=c( -5,-5,-15,-25,-25,-35),
y=c( -5, 5, 0, 0, -5, 5)
),
places_df=data.frame(id=1:4,
name=c("bgn","fit.data.training.all","predict.data.new","end"),
x=c( -0, -20, -30, -40),
y=c( 0, 0, 0, 0),
M0=c( 3, 0, 0, 0)
),
arcs_df = data.frame(
begin = c("bgn","bgn","bgn",
"data.training.all","model.selected","fit.data.training.all",
"fit.data.training.all","model.final",
"data.new","predict.data.new",
"data.training.all.prediction","data.new.prediction"),
end = c("data.training.all","data.new","model.selected",
"fit.data.training.all","fit.data.training.all","model.final",
"data.training.all.prediction","predict.data.new",
"predict.data.new","data.new.prediction",
"end","end")
))
#print(ggplot.petrinet(glb_analytics_pn))
print(ggplot.petrinet(glb_analytics_pn) + coord_flip())## Loading required package: grid
glb_analytics_avl_objs <- NULL
glb_chunks_df <- myadd_chunk(NULL, "import.data")## label step_major step_minor label_minor bgn end elapsed
## 1 import.data 1 0 0 24.223 NA NAStep 1.0: import data
chunk option: eval=
## [1] "Reading file ./data/train_resXY.csv..."
## [1] "dimensions of data in ./data/train_resXY.csv: 2,000 rows x 5 cols"
## [1] " Truncating resXLst to first 100 chars..."
## [1] " Truncating resYLst to first 100 chars..."
## business_id labels nImgs
## 1 1000 1 2 3 4 5 6 7 54
## 2 1001 0 1 6 8 9
## 3 100 1 2 4 5 6 7 84
## 4 1006 1 2 4 5 6 22
## 5 1010 0 6 8 11
## 6 101 1 2 3 4 5 6 121
## resXLst
## 1 500,375,375,375,375,375,500,500,500,500,500,500,500,500,375,414,373,500,399,375,375,375,500,500,472,
## 2 500,375,500,500,500,366,358,444,500
## 3 500,375,375,375,375,500,375,375,500,375,373,375,375,500,375,500,500,500,500,375,375,375,375,375,375,
## 4 500,373,281,500,500,500,500,500,500,500,500,396,500,500,500,281,281,375,375,375,375,375
## 5 375,500,375,500,500,500,500,375,500,500,500
## 6 375,299,299,299,299,299,299,373,373,373,373,500,500,408,500,500,500,500,375,500,373,500,500,375,375,
## resYLst
## 1 500,500,500,500,500,500,332,332,332,332,332,375,375,375,500,500,500,389,500,500,500,500,375,375,500,
## 2 375,500,375,361,375,500,500,479,373
## 3 375,500,500,500,500,375,500,500,268,500,500,500,500,375,500,375,375,375,375,500,500,500,500,500,500,
## 4 375,500,500,273,375,375,375,375,375,399,290,500,500,500,375,500,500,500,500,500,500,500
## 5 500,375,500,375,375,375,375,500,375,375,375
## 6 500,500,500,500,500,500,500,500,500,500,500,282,282,306,388,375,375,375,500,373,500,348,386,500,500,
## business_id labels nImgs
## 69 1102 6 8 37
## 305 1479 0 3 8 306
## 1019 2829 0 2 3 8 104
## 1455 3650 8 42
## 1468 3675 1 2 3 4 5 6 7 32
## 1978 959 3 5 6 8 29
## resXLst
## 69 375,500,375,500,375,373,500,452,468,500,500,500,500,500,375,500,500,500,500,500,500,500,373,500,500,
## 305 500,373,373,500,500,500,500,373,375,500,500,500,500,373,373,373,375,500,375,373,373,375,373,500,500,
## 1019 500,375,375,500,500,375,375,375,500,373,375,375,500,375,500,375,375,375,374,375,500,375,375,500,375,
## 1455 375,375,375,375,375,375,375,500,375,500,500,375,375,375,500,375,500,375,500,373,375,500,500,375,281,
## 1468 156,375,500,500,500,500,500,500,500,375,375,375,500,500,500,375,500,375,375,500,500,375,441,500,433,
## 1978 500,500,375,500,500,374,373,373,373,500,500,373,500,375,500,500,375,375,500,375,375,500,375,500,500,
## resYLst
## 69 500,375,500,375,500,500,373,500,500,375,375,375,375,332,500,375,375,375,375,375,375,373,500,373,500,
## 305 373,500,500,373,500,500,500,500,500,375,375,375,375,500,500,500,500,299,500,500,500,500,500,331,375,
## 1019 500,500,500,375,375,500,500,500,375,500,500,500,375,500,375,500,500,500,500,500,500,500,500,375,500,
## 1455 500,500,500,500,500,500,500,375,500,375,281,500,500,500,376,500,376,500,375,500,500,376,376,500,500,
## 1468 121,500,340,500,375,375,375,375,375,500,500,500,375,375,375,500,375,500,500,375,283,500,500,294,500,
## 1978 373,373,500,375,375,500,500,500,500,375,373,500,281,500,375,375,500,500,375,500,500,373,500,375,375,
## business_id labels nImgs
## 1995 99 1 2 4 5 6 7 139
## 1996 991 1 2 3 5 6 7 84
## 1997 993 3 6 8 34
## 1998 997 8 107
## 1999 998 1 2 4 5 6 7 320
## 2000 999 1 2 5 6 7 33
## resXLst
## 1995 500,500,500,500,373,373,500,500,500,500,500,500,500,500,375,500,500,375,500,375,375,375,375,500,375,
## 1996 500,373,500,375,375,375,375,375,375,375,415,500,500,352,480,375,500,500,281,500,500,500,500,500,500,
## 1997 500,500,500,500,500,299,500,500,375,500,500,500,375,375,375,375,375,500,500,500,375,500,500,373,500,
## 1998 320,376,375,500,500,500,467,467,500,500,500,500,500,500,500,500,500,500,500,500,282,375,500,500,500,
## 1999 500,500,500,375,500,500,500,500,500,375,500,500,500,500,500,375,500,500,500,500,375,375,375,375,375,
## 2000 375,500,500,500,500,500,500,500,375,375,375,299,232,500,500,500,500,500,500,500,500,500,375,500,500,
## resYLst
## 1995 375,500,500,500,500,500,375,375,375,375,375,375,375,375,500,375,375,500,375,500,500,500,500,500,500,
## 1996 500,500,373,500,500,500,500,500,500,500,499,387,375,500,360,500,375,500,500,375,373,373,373,373,373,
## 1997 373,375,375,375,299,500,299,375,500,375,375,375,500,500,500,500,500,373,280,280,500,281,373,500,373,
## 1998 240,500,500,414,375,375,351,351,373,373,373,336,336,336,336,343,336,337,337,343,500,500,500,500,500,
## 1999 450,279,281,500,375,375,373,282,375,500,375,340,375,375,375,500,373,373,373,373,500,500,500,500,500,
## 2000 500,375,375,375,375,375,375,375,500,500,500,500,64,281,281,281,375,375,375,375,375,375,500,500,500,5
## 'data.frame': 2000 obs. of 5 variables:
## $ business_id: int 1000 1001 100 1006 1010 101 1011 1012 1014 1015 ...
## $ labels : chr "1 2 3 4 5 6 7" "0 1 6 8" "1 2 4 5 6 7" "1 2 4 5 6" ...
## $ nImgs : int 54 9 84 22 11 121 70 37 32 145 ...
## $ resXLst : chr "500,375,375,375,375,375,500,500,500,500,500,500,500,500,375,414,373,500,399,375,375,375,500,500,472,478,467,470,375,373,375,375"| __truncated__ "500,375,500,500,500,366,358,444,500" "500,375,375,375,375,500,375,375,500,375,373,375,375,500,375,500,500,500,500,375,375,375,375,375,375,375,375,373,373,375,375,375"| __truncated__ "500,373,281,500,500,500,500,500,500,500,500,396,500,500,500,281,281,375,375,375,375,375" ...
## $ resYLst : chr "500,500,500,500,500,500,332,332,332,332,332,375,375,375,500,500,500,389,500,500,500,500,375,375,500,500,500,499,500,500,500,500"| __truncated__ "375,500,375,361,375,500,500,479,373" "375,500,500,500,500,375,500,500,268,500,500,500,500,375,500,375,375,375,375,500,500,500,500,500,500,500,500,500,500,500,500,500"| __truncated__ "375,500,500,273,375,375,375,375,375,399,290,500,500,500,375,500,500,500,500,500,500,500" ...
## - attr(*, "comment")= chr "glbObsTrn"
## NULL## [1] "Reading file ./data/test_resXY.csv..."
## [1] "dimensions of data in ./data/test_resXY.csv: 10,000 rows x 4 cols"
## [1] " Truncating resXLst to first 100 chars..."
## [1] " Truncating resYLst to first 100 chars..."
## business_id nImgs
## 1 003sg 167
## 2 00er5 210
## 3 00kad 83
## 4 00mc6 15
## 5 00q7x 24
## 6 00v0t 24
## resXLst
## 1 375,500,375,375,500,375,500,500,500,500,500,281,500,500,500,373,375,500,375,373,375,500,500,375,500,
## 2 489,500,500,281,375,397,469,500,320,375,500,375,500,375,375,375,500,500,345,375,375,500,500,500,281,
## 3 332,500,500,375,281,375,500,500,500,375,500,500,375,375,375,500,500,500,500,375,500,500,375,500,500,
## 4 375,500,500,375,323,500,500,500,281,500,375,500,500,500,375
## 5 500,500,373,375,500,500,500,500,375,500,375,375,325,500,375,500,500,500,500,375,500,375,375,500
## 6 375,500,375,500,375,500,500,500,500,373,500,375,375,375,500,500,500,500,375,500,500,500,375,375
## resYLst
## 1 500,375,500,500,500,500,375,375,350,375,500,500,375,375,373,500,500,500,500,500,500,332,375,500,375,
## 2 500,500,375,500,500,500,314,375,480,500,375,500,375,500,500,500,373,375,500,500,500,375,380,282,500,
## 3 500,375,375,500,500,500,281,281,281,500,375,500,500,500,500,375,375,374,335,500,373,344,500,373,375,
## 4 500,500,375,500,500,375,418,375,500,375,500,375,500,333,500
## 5 375,500,500,500,375,375,372,333,500,375,500,500,500,375,500,375,375,375,375,500,373,500,500,500
## 6 500,375,500,375,500,375,375,375,281,500,375,500,500,500,500,375,375,375,500,500,375,282,500,500
## business_id nImgs
## 12 01mrb 62
## 1789 6ey8p 40
## 3881 dqqme 117
## 3912 dv9lg 15
## 4024 ebyno 128
## 4625 gkb3z 44
## resXLst
## 12 500,500,500,500,500,375,500,500,373,500,500,500,375,375,375,333,500,375,375,500,500,375,332,500,464,
## 1789 500,373,375,500,500,375,500,500,375,500,375,500,360,500,500,500,375,375,500,500,500,500,500,350,500,
## 3881 500,500,375,500,500,375,500,375,375,500,375,500,500,281,375,375,376,500,500,375,375,500,500,281,375,
## 3912 281,500,500,281,375,500,500,362,500,500,375,500,500,281,500
## 4024 375,375,375,500,372,375,500,500,373,500,375,375,500,500,373,282,375,500,500,281,375,375,500,500,375,
## 4625 375,500,500,500,375,500,500,500,500,500,500,500,375,500,500,500,373,375,375,375,375,500,500,375,500,
## resYLst
## 12 375,333,375,375,375,500,375,375,500,375,375,334,500,500,500,500,375,500,500,500,375,500,500,375,368,
## 1789 375,500,500,333,500,500,373,500,500,333,500,375,450,375,375,375,500,500,373,375,374,375,375,263,373,
## 3881 334,433,500,375,375,500,375,500,500,500,500,299,375,500,500,500,500,375,375,500,500,375,500,500,500,
## 3912 500,375,375,500,500,375,373,500,375,375,500,375,375,500,375
## 4024 500,500,500,375,500,500,281,500,500,375,500,500,500,319,500,500,500,375,375,500,500,500,375,375,500,
## 4625 500,375,375,375,500,375,375,375,373,469,373,373,500,442,413,373,500,500,500,500,500,375,375,500,375,
## business_id nImgs
## 9995 zyrif 89
## 9996 zyvg6 16
## 9997 zyvjj 27
## 9998 zz8g4 118
## 9999 zzxkg 154
## 10000 zzxwm 13
## resXLst
## 9995 375,500,375,500,500,500,375,375,500,500,375,375,500,375,375,500,500,281,281,500,500,375,375,500,500,
## 9996 500,500,375,373,500,500,500,375,500,375,500,375,280,375,500,375
## 9997 500,375,500,500,500,500,500,402,500,373,500,375,500,500,500,500,375,500,500,375,500,375,500,500,281,
## 9998 375,500,375,500,375,375,375,500,500,375,500,500,500,500,500,499,500,500,500,375,282,500,500,500,375,
## 9999 500,500,500,500,375,500,500,500,375,375,375,299,500,500,375,500,500,375,500,500,500,373,500,281,500,
## 10000 500,373,500,281,500,375,333,375,375,218,500,500,500
## resYLst
## 9995 500,375,500,375,375,375,500,500,500,281,500,500,375,500,500,375,375,500,500,373,375,500,500,500,373,
## 9996 375,375,500,500,375,500,375,500,375,500,375,500,500,500,375,500
## 9997 373,500,375,375,406,373,373,315,373,500,375,500,281,373,375,375,500,280,373,500,375,500,375,375,500,
## 9998 500,375,500,232,500,500,500,375,375,500,375,375,375,373,500,323,334,373,375,500,500,335,375,280,500,
## 9999 375,375,375,375,500,373,375,375,500,500,500,500,375,500,500,375,371,500,281,375,375,500,375,500,375,
## 10000 281,500,299,500,374,500,500,500,500,211,375,375,281
## 'data.frame': 10000 obs. of 4 variables:
## $ business_id: chr "003sg" "00er5" "00kad" "00mc6" ...
## $ nImgs : int 167 210 83 15 24 24 40 10 49 10 ...
## $ resXLst : chr "375,500,375,375,500,375,500,500,500,500,500,281,500,500,500,373,375,500,375,373,375,500,500,375,500,500,375,279,500,375,500,500"| __truncated__ "489,500,500,281,375,397,469,500,320,375,500,375,500,375,375,375,500,500,345,375,375,500,500,500,281,373,500,375,500,375,500,375"| __truncated__ "332,500,500,375,281,375,500,500,500,375,500,500,375,375,375,500,500,500,500,375,500,500,375,500,500,500,500,432,281,373,500,297"| __truncated__ "375,500,500,375,323,500,500,500,281,500,375,500,500,500,375" ...
## $ resYLst : chr "500,375,500,500,500,500,375,375,350,375,500,500,375,375,373,500,500,500,500,500,500,332,375,500,375,375,500,500,375,500,375,500"| __truncated__ "500,500,375,500,500,500,314,375,480,500,375,500,375,500,500,500,373,375,500,500,500,375,380,282,500,500,281,500,361,500,375,500"| __truncated__ "500,375,375,500,500,500,281,281,281,500,375,500,500,500,500,375,375,374,335,500,373,344,500,373,375,375,333,500,500,500,299,500"| __truncated__ "500,500,375,500,500,375,418,375,500,375,500,375,500,333,500" ...
## - attr(*, "comment")= chr "glbObsNew"
## NULL## [1] "Creating new feature: .pos..."
## [1] "Creating new feature: nImgs.log1p..."
## [1] "Creating new feature: nImgs.root2..."
## [1] "Creating new feature: nImgs.nexp..."
## [1] "Creating new feature: resX.min..."
## [1] "Creating new feature: resX.max..."
## [1] "Creating new feature: resX.mean..."
## [1] "Creating new feature: resX.mad..."
## [1] "Creating new feature: resX.min.log1p..."
## [1] "Creating new feature: resX.min.root2..."
## [1] "Creating new feature: resX.min.nexp..."
## [1] "Creating new feature: resX.max.log1p..."
## [1] "Creating new feature: resX.max.root2..."
## [1] "Creating new feature: resX.max.nexp..."
## [1] "Creating new feature: resX.mean.log1p..."
## [1] "Creating new feature: resX.mean.root2..."
## [1] "Creating new feature: resX.mean.nexp..."
## [1] "Creating new feature: resX.mad.log1p..."
## [1] "Creating new feature: resX.mad.root2..."
## [1] "Creating new feature: resX.mad.nexp..."
## [1] "Creating new feature: resY.min..."
## [1] "Creating new feature: resY.max..."
## [1] "Creating new feature: resY.mean..."
## [1] "Creating new feature: resY.mad..."
## [1] "Creating new feature: resY.min.log1p..."
## [1] "Creating new feature: resY.min.root2..."
## [1] "Creating new feature: resY.min.nexp..."
## [1] "Creating new feature: resY.max.log1p..."
## [1] "Creating new feature: resY.max.root2..."
## [1] "Creating new feature: resY.max.nexp..."
## [1] "Creating new feature: resY.mean.log1p..."
## [1] "Creating new feature: resY.mean.root2..."
## [1] "Creating new feature: resY.mean.nexp..."
## [1] "Creating new feature: resY.mad.log1p..."
## [1] "Creating new feature: resY.mad.root2..."
## [1] "Creating new feature: resY.mad.nexp..."
## [1] "Creating new feature: resXY.min..."
## [1] "Creating new feature: resXY.max..."
## [1] "Creating new feature: resXY.mean..."
## [1] "Creating new feature: resXY.mad..."
## [1] "Creating new feature: resXY.min.log1p..."
## [1] "Creating new feature: resXY.min.root2..."
## [1] "Creating new feature: resXY.min.nexp..."
## [1] "Creating new feature: resXY.max.log1p..."
## [1] "Creating new feature: resXY.max.root2..."
## [1] "Creating new feature: resXY.max.nexp..."
## [1] "Creating new feature: resXY.mean.log1p..."
## [1] "Creating new feature: resXY.mean.root2..."
## [1] "Creating new feature: resXY.mean.nexp..."
## [1] "Creating new feature: resXY.mad.log1p..."
## [1] "Creating new feature: resXY.mad.root2..."
## [1] "Creating new feature: resXY.mad.nexp..."
## [1] "Creating new feature: lunch..."
## [1] "Creating new feature: dinner..."
## [1] "Creating new feature: reserve..."
## [1] "Creating new feature: outdoor..."
## [1] "Creating new feature: expensive..."
## [1] "Creating new feature: liquor..."
## [1] "Creating new feature: table..."
## [1] "Creating new feature: classy..."
## [1] "Creating new feature: kids..."
## [1] "Creating new feature: nImgs.cut.fctr..."## [1] "Partition stats:"## Loading required package: sqldf## Loading required package: gsubfn## Loading required package: proto## Loading required package: RSQLite## Loading required package: DBI## Loading required package: tcltk## outdoor .src .n
## 1 <NA> Test 10000
## 2 3 Train 1003
## 3 -1 Train 997
## outdoor .src .n
## 1 <NA> Test 10000
## 2 3 Train 1003
## 3 -1 Train 997## Loading required package: RColorBrewer
## .src .n
## 1 Test 10000
## 2 Train 2000## Loading required package: lazyeval## Loading required package: gdata## gdata: read.xls support for 'XLS' (Excel 97-2004) files ENABLED.## ## gdata: read.xls support for 'XLSX' (Excel 2007+) files ENABLED.##
## Attaching package: 'gdata'## The following objects are masked from 'package:dplyr':
##
## combine, first, last## The following object is masked from 'package:stats':
##
## nobs## The following object is masked from 'package:utils':
##
## object.size## [1] "Found 0 duplicates by all features:"## NULL## label step_major step_minor label_minor bgn end elapsed
## 1 import.data 1 0 0 24.223 79.635 55.412
## 2 inspect.data 2 0 0 79.636 NA NAStep 2.0: inspect data
## Loading required package: reshape2
## outdoor.-1 outdoor.3 outdoor.NA
## Test NA NA 10000
## Train 997 1003 NA
## outdoor.-1 outdoor.3 outdoor.NA
## Test NA NA 1
## Train 0.4985 0.5015 NA
## [1] "numeric data missing in glbObsAll: "
## lunch dinner reserve outdoor expensive liquor table
## 10000 10000 10000 10000 10000 10000 10000
## classy kids
## 10000 10000
## [1] "numeric data w/ 0s in glbObsAll: "
## nImgs.nexp resX.mad resX.mad.log1p resX.mad.root2
## 228 9353 9353 9353
## resY.mad resY.mad.log1p resY.mad.root2 resXY.mad
## 5442 5442 5442 10915
## resXY.min.nexp resXY.max.nexp resXY.mean.nexp resXY.mad.log1p
## 12000 12000 12000 10915
## resXY.mad.root2 resXY.mad.nexp lunch
## 10915 850 671
## [1] "numeric data w/ Infs in glbObsAll: "
## named integer(0)
## [1] "numeric data w/ NaNs in glbObsAll: "
## named integer(0)
## [1] "string data missing in glbObsAll: "
## business_id labels resXLst resYLst
## 0 NA 0 0## outdoor outdoor.fctr .n
## 1 <NA> <NA> 10000
## 2 3 Y 1003
## 3 -1 N 997## Warning: Removed 1 rows containing missing values (position_stack).
## outdoor.fctr.N outdoor.fctr.Y outdoor.fctr.NA
## Test NA NA 10000
## Train 997 1003 NA
## outdoor.fctr.N outdoor.fctr.Y outdoor.fctr.NA
## Test NA NA 1
## Train 0.4985 0.5015 NA
## NULL## label step_major step_minor label_minor bgn end elapsed
## 2 inspect.data 2 0 0 79.636 116.109 36.474
## 3 scrub.data 2 1 1 116.110 NA NAStep 2.1: scrub data
## [1] "numeric data missing in glbObsAll: "
## lunch dinner reserve outdoor expensive
## 10000 10000 10000 10000 10000
## liquor table classy kids outdoor.fctr
## 10000 10000 10000 10000 10000
## [1] "numeric data w/ 0s in glbObsAll: "
## nImgs.nexp resX.mad resX.mad.log1p resX.mad.root2
## 228 9353 9353 9353
## resY.mad resY.mad.log1p resY.mad.root2 resXY.mad
## 5442 5442 5442 10915
## resXY.min.nexp resXY.max.nexp resXY.mean.nexp resXY.mad.log1p
## 12000 12000 12000 10915
## resXY.mad.root2 resXY.mad.nexp lunch
## 10915 850 671
## [1] "numeric data w/ Infs in glbObsAll: "
## named integer(0)
## [1] "numeric data w/ NaNs in glbObsAll: "
## named integer(0)
## [1] "string data missing in glbObsAll: "
## business_id labels resXLst resYLst
## 0 NA 0 0## label step_major step_minor label_minor bgn end elapsed
## 3 scrub.data 2 1 1 116.110 126.864 10.754
## 4 transform.data 2 2 2 126.865 NA NAStep 2.2: transform data
## label step_major step_minor label_minor bgn end
## 4 transform.data 2 2 2 126.865 126.905
## 5 extract.features 3 0 0 126.905 NA
## elapsed
## 4 0.04
## 5 NAStep 3.0: extract features
## label step_major step_minor label_minor bgn
## 5 extract.features 3 0 0 126.905
## 6 extract.features.datetime 3 1 1 126.926
## end elapsed
## 5 126.926 0.021
## 6 NA NAStep 3.1: extract features datetime
## label step_major step_minor label_minor bgn
## 1 extract.features.datetime.bgn 1 0 0 126.977
## end elapsed
## 1 NA NA## label step_major step_minor label_minor bgn
## 6 extract.features.datetime 3 1 1 126.926
## 7 extract.features.image 3 2 2 126.988
## end elapsed
## 6 126.987 0.061
## 7 NA NAStep 3.2: extract features image
```{r extract.features.image, cache=FALSE, echo=FALSE, fig.height=5, fig.width=5, eval=myevlChunk(glbChunks, glbOut$pfx)}
## label step_major step_minor label_minor bgn end
## 1 extract.features.image.bgn 1 0 0 127.029 NA
## elapsed
## 1 NA## label step_major step_minor label_minor bgn
## 1 extract.features.image.bgn 1 0 0 127.029
## 2 extract.features.image.end 2 0 0 127.039
## end elapsed
## 1 127.038 0.01
## 2 NA NA## label step_major step_minor label_minor bgn
## 1 extract.features.image.bgn 1 0 0 127.029
## 2 extract.features.image.end 2 0 0 127.039
## end elapsed
## 1 127.038 0.01
## 2 NA NA## label step_major step_minor label_minor bgn end
## 7 extract.features.image 3 2 2 126.988 127.048
## 8 extract.features.price 3 3 3 127.049 NA
## elapsed
## 7 0.06
## 8 NAStep 3.3: extract features price
## label step_major step_minor label_minor bgn end
## 1 extract.features.price.bgn 1 0 0 127.074 NA
## elapsed
## 1 NA## label step_major step_minor label_minor bgn end
## 8 extract.features.price 3 3 3 127.049 127.083
## 9 extract.features.text 3 4 4 127.083 NA
## elapsed
## 8 0.034
## 9 NAStep 3.4: extract features text
## label step_major step_minor label_minor bgn end
## 1 extract.features.text.bgn 1 0 0 127.14 NA
## elapsed
## 1 NA## label step_major step_minor label_minor bgn
## 9 extract.features.text 3 4 4 127.083
## 10 extract.features.string 3 5 5 127.150
## end elapsed
## 9 127.149 0.066
## 10 NA NAStep 3.5: extract features string
## label step_major step_minor label_minor bgn end
## 1 extract.features.string.bgn 1 0 0 127.18 NA
## elapsed
## 1 NA## label step_major step_minor
## 1 extract.features.string.bgn 1 0
## 2 extract.features.stringfactorize.str.vars 2 0
## label_minor bgn end elapsed
## 1 0 127.180 127.189 0.009
## 2 0 127.189 NA NA## business_id labels resXLst resYLst .src
## "business_id" "labels" "resXLst" "resYLst" ".src"## label step_major step_minor label_minor bgn
## 10 extract.features.string 3 5 5 127.150
## 11 extract.features.end 3 6 6 127.205
## end elapsed
## 10 127.204 0.054
## 11 NA NAStep 3.6: extract features end
## [1] "Summary for lunch:"
##
## -1 0 <NA>
## Test 0 0 10000
## Train 1329 671 0
## [1] "Summary for dinner:"
##
## -1 1 <NA>
## Test 0 0 10000
## Train 1007 993 0
## [1] "Summary for reserve:"
##
## -1 2 <NA>
## Test 0 0 10000
## Train 974 1026 0
## [1] "Summary for outdoor:"
##
## -1 3 <NA>
## Test 0 0 10000
## Train 997 1003 0
## [1] "Summary for expensive:"
##
## -1 4 <NA>
## Test 0 0 10000
## Train 1453 547 0
## [1] "Summary for liquor:"
##
## -1 5 <NA>
## Test 0 0 10000
## Train 751 1249 0
## [1] "Summary for table:"
##
## -1 6 <NA>
## Test 0 0 10000
## Train 640 1360 0
## [1] "Summary for classy:"
##
## -1 7 <NA>
## Test 0 0 10000
## Train 1428 572 0
## [1] "Summary for kids:"
##
## -1 8 <NA>
## Test 0 0 10000
## Train 762 1238 0## time trans "bgn " "fit.data.training.all " "predict.data.new " "end "
## 0.0000 multiple enabled transitions: data.training.all data.new model.selected firing: data.training.all
## 1.0000 1 2 1 0 0
## 1.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction firing: data.new
## 2.0000 2 1 1 1 0
## label step_major step_minor label_minor bgn end
## 11 extract.features.end 3 6 6 127.205 128.134
## 12 manage.missing.data 4 0 0 128.135 NA
## elapsed
## 11 0.929
## 12 NAStep 4.0: manage missing data
## [1] "numeric data missing in glbObsAll: "
## lunch dinner reserve outdoor expensive
## 10000 10000 10000 10000 10000
## liquor table classy kids outdoor.fctr
## 10000 10000 10000 10000 10000
## [1] "numeric data w/ 0s in glbObsAll: "
## nImgs.nexp resX.mad resX.mad.log1p resX.mad.root2
## 228 9353 9353 9353
## resY.mad resY.mad.log1p resY.mad.root2 resXY.mad
## 5442 5442 5442 10915
## resXY.min.nexp resXY.max.nexp resXY.mean.nexp resXY.mad.log1p
## 12000 12000 12000 10915
## resXY.mad.root2 resXY.mad.nexp lunch
## 10915 850 671
## [1] "numeric data w/ Infs in glbObsAll: "
## named integer(0)
## [1] "numeric data w/ NaNs in glbObsAll: "
## named integer(0)
## [1] "string data missing in glbObsAll: "
## business_id labels resXLst resYLst
## 0 NA 0 0## [1] "numeric data missing in glbObsAll: "
## lunch dinner reserve outdoor expensive
## 10000 10000 10000 10000 10000
## liquor table classy kids outdoor.fctr
## 10000 10000 10000 10000 10000
## [1] "numeric data w/ 0s in glbObsAll: "
## nImgs.nexp resX.mad resX.mad.log1p resX.mad.root2
## 228 9353 9353 9353
## resY.mad resY.mad.log1p resY.mad.root2 resXY.mad
## 5442 5442 5442 10915
## resXY.min.nexp resXY.max.nexp resXY.mean.nexp resXY.mad.log1p
## 12000 12000 12000 10915
## resXY.mad.root2 resXY.mad.nexp lunch
## 10915 850 671
## [1] "numeric data w/ Infs in glbObsAll: "
## named integer(0)
## [1] "numeric data w/ NaNs in glbObsAll: "
## named integer(0)
## [1] "string data missing in glbObsAll: "
## business_id labels resXLst resYLst
## 0 NA 0 0## label step_major step_minor label_minor bgn end
## 12 manage.missing.data 4 0 0 128.135 128.566
## 13 cluster.data 5 0 0 128.567 NA
## elapsed
## 12 0.431
## 13 NAStep 5.0: cluster data
## label step_major step_minor label_minor bgn
## 13 cluster.data 5 0 0 128.567
## 14 partition.data.training 6 0 0 128.635
## end elapsed
## 13 128.635 0.068
## 14 NA NAStep 6.0: partition data training
## [1] "partition.data.training chunk: setup: elapsed: 0.00 secs"## [1] "partition.data.training chunk: strata_mtrx complete: elapsed: 0.15 secs"
## [1] "partition.data.training chunk: obs_freq_df complete: elapsed: 0.15 secs"## Loading required package: sampling##
## Attaching package: 'sampling'## The following objects are masked from 'package:survival':
##
## cluster, strata## The following object is masked from 'package:caret':
##
## cluster## [1] "partition.data.training chunk: Fit/OOB partition complete: elapsed: 0.41 secs"## outdoor.-1 outdoor.3 outdoor.NA
## NA NA 10000
## Fit 500 502 NA
## OOB 497 501 NA
## outdoor.-1 outdoor.3 outdoor.NA
## NA NA 1
## Fit 0.499002 0.500998 NA
## OOB 0.497996 0.502004 NA
## nImgs.cut.fctr .n.Fit .n.OOB .n.Tst .freqRatio.Fit .freqRatio.OOB
## 1 (0,32] 238 237 2532 0.2375250 0.2374749
## 3 (32,60] 278 243 2512 0.2774451 0.2434870
## 2 (120,3e+03] 229 258 2497 0.2285429 0.2585170
## 4 (60,120] 257 260 2459 0.2564870 0.2605210
## .freqRatio.Tst
## 1 0.2532
## 3 0.2512
## 2 0.2497
## 4 0.2459## [1] "glbObsAll: "## [1] 12000 71## [1] "glbObsTrn: "## [1] 2000 71## [1] "glbObsFit: "## [1] 1002 70## [1] "glbObsOOB: "## [1] 998 70## [1] "glbObsNew: "## [1] 10000 70## [1] "partition.data.training chunk: teardown: elapsed: 1.23 secs"## label step_major step_minor label_minor bgn
## 14 partition.data.training 6 0 0 128.635
## 15 select.features 7 0 0 129.924
## end elapsed
## 14 129.923 1.289
## 15 NA NAStep 7.0: select features
## Warning in cor(data.matrix(entity_df[, sel_feats]), y =
## as.numeric(entity_df[, : the standard deviation is zero## [1] "cor(resX.max, resX.max.log1p)=1.0000"
## [1] "cor(outdoor.fctr, resX.max)=-0.0315"
## [1] "cor(outdoor.fctr, resX.max.log1p)=-0.0315"## Warning in myfind_cor_features(feats_df = glb_feats_df, obs_df =
## glbObsTrn, : Identified resX.max.log1p as highly correlated with resX.max## [1] "cor(resX.max, resX.max.nexp)=-1.0000"
## [1] "cor(outdoor.fctr, resX.max)=-0.0315"
## [1] "cor(outdoor.fctr, resX.max.nexp)=0.0315"## Warning in myfind_cor_features(feats_df = glb_feats_df, obs_df =
## glbObsTrn, : Identified resX.max as highly correlated with resX.max.nexp## [1] "cor(resX.max.nexp, resX.max.root2)=-1.0000"
## [1] "cor(outdoor.fctr, resX.max.nexp)=0.0315"
## [1] "cor(outdoor.fctr, resX.max.root2)=-0.0315"## Warning in myfind_cor_features(feats_df = glb_feats_df, obs_df =
## glbObsTrn, : Identified resX.max.root2 as highly correlated with
## resX.max.nexp## [1] "cor(resX.mean.nexp, resY.mean.nexp)=1.0000"
## [1] "cor(outdoor.fctr, resX.mean.nexp)=-0.0224"
## [1] "cor(outdoor.fctr, resY.mean.nexp)=-0.0224"## Warning in myfind_cor_features(feats_df = glb_feats_df, obs_df =
## glbObsTrn, : Identified resY.mean.nexp as highly correlated with
## resX.mean.nexp## [1] "cor(resY.max, resY.max.root2)=1.0000"
## [1] "cor(outdoor.fctr, resY.max)=0.0117"
## [1] "cor(outdoor.fctr, resY.max.root2)=0.0116"## Warning in myfind_cor_features(feats_df = glb_feats_df, obs_df =
## glbObsTrn, : Identified resY.max.root2 as highly correlated with resY.max## [1] "cor(resY.max, resY.max.log1p)=1.0000"
## [1] "cor(outdoor.fctr, resY.max)=0.0117"
## [1] "cor(outdoor.fctr, resY.max.log1p)=0.0115"## Warning in myfind_cor_features(feats_df = glb_feats_df, obs_df =
## glbObsTrn, : Identified resY.max.log1p as highly correlated with resY.max## [1] "cor(resX.mean, resX.mean.root2)=0.9996"
## [1] "cor(outdoor.fctr, resX.mean)=-0.0177"
## [1] "cor(outdoor.fctr, resX.mean.root2)=-0.0164"## Warning in myfind_cor_features(feats_df = glb_feats_df, obs_df =
## glbObsTrn, : Identified resX.mean.root2 as highly correlated with resX.mean## [1] "cor(resY.mean, resY.mean.root2)=0.9995"
## [1] "cor(outdoor.fctr, resY.mean)=0.0126"
## [1] "cor(outdoor.fctr, resY.mean.root2)=0.0131"## Warning in myfind_cor_features(feats_df = glb_feats_df, obs_df =
## glbObsTrn, : Identified resY.mean as highly correlated with resY.mean.root2## [1] "cor(resY.mean.log1p, resY.mean.root2)=0.9994"
## [1] "cor(outdoor.fctr, resY.mean.log1p)=0.0136"
## [1] "cor(outdoor.fctr, resY.mean.root2)=0.0131"## Warning in myfind_cor_features(feats_df = glb_feats_df, obs_df =
## glbObsTrn, : Identified resY.mean.root2 as highly correlated with
## resY.mean.log1p## [1] "cor(resX.mean, resX.mean.log1p)=0.9985"
## [1] "cor(outdoor.fctr, resX.mean)=-0.0177"
## [1] "cor(outdoor.fctr, resX.mean.log1p)=-0.0151"## Warning in myfind_cor_features(feats_df = glb_feats_df, obs_df =
## glbObsTrn, : Identified resX.mean.log1p as highly correlated with resX.mean## [1] "cor(resX.min, resX.min.root2)=0.9940"
## [1] "cor(outdoor.fctr, resX.min)=-0.0314"
## [1] "cor(outdoor.fctr, resX.min.root2)=-0.0303"## Warning in myfind_cor_features(feats_df = glb_feats_df, obs_df =
## glbObsTrn, : Identified resX.min.root2 as highly correlated with resX.min## [1] "cor(resY.min, resY.min.root2)=0.9935"
## [1] "cor(outdoor.fctr, resY.min)=-0.0509"
## [1] "cor(outdoor.fctr, resY.min.root2)=-0.0474"## Warning in myfind_cor_features(feats_df = glb_feats_df, obs_df =
## glbObsTrn, : Identified resY.min.root2 as highly correlated with resY.min## [1] "cor(resX.mad.log1p, resX.mad.root2)=0.9880"
## [1] "cor(outdoor.fctr, resX.mad.log1p)=0.0220"
## [1] "cor(outdoor.fctr, resX.mad.root2)=0.0219"## Warning in myfind_cor_features(feats_df = glb_feats_df, obs_df =
## glbObsTrn, : Identified resX.mad.root2 as highly correlated with
## resX.mad.log1p## [1] "cor(resXY.min, resXY.min.root2)=0.9872"
## [1] "cor(outdoor.fctr, resXY.min)=-0.0495"
## [1] "cor(outdoor.fctr, resXY.min.root2)=-0.0414"## Warning in myfind_cor_features(feats_df = glb_feats_df, obs_df =
## glbObsTrn, : Identified resXY.min.root2 as highly correlated with resXY.min## [1] "cor(resXY.mad.log1p, resXY.mad.nexp)=-0.9803"
## [1] "cor(outdoor.fctr, resXY.mad.log1p)=-0.0141"
## [1] "cor(outdoor.fctr, resXY.mad.nexp)=0.0154"## Warning in myfind_cor_features(feats_df = glb_feats_df, obs_df =
## glbObsTrn, : Identified resXY.mad.log1p as highly correlated with
## resXY.mad.nexp## [1] "cor(resX.min, resX.min.log1p)=0.9751"
## [1] "cor(outdoor.fctr, resX.min)=-0.0314"
## [1] "cor(outdoor.fctr, resX.min.log1p)=-0.0301"## Warning in myfind_cor_features(feats_df = glb_feats_df, obs_df =
## glbObsTrn, : Identified resX.min.log1p as highly correlated with resX.min## [1] "cor(resY.min, resY.min.log1p)=0.9718"
## [1] "cor(outdoor.fctr, resY.min)=-0.0509"
## [1] "cor(outdoor.fctr, resY.min.log1p)=-0.0431"## Warning in myfind_cor_features(feats_df = glb_feats_df, obs_df =
## glbObsTrn, : Identified resY.min.log1p as highly correlated with resY.min## [1] "cor(resX.mad, resX.mad.log1p)=0.9375"
## [1] "cor(outdoor.fctr, resX.mad)=0.0205"
## [1] "cor(outdoor.fctr, resX.mad.log1p)=0.0220"## Warning in myfind_cor_features(feats_df = glb_feats_df, obs_df =
## glbObsTrn, : Identified resX.mad as highly correlated with resX.mad.log1p## [1] "cor(resXY.min, resXY.min.log1p)=0.9357"
## [1] "cor(outdoor.fctr, resXY.min)=-0.0495"
## [1] "cor(outdoor.fctr, resXY.min.log1p)=-0.0338"## Warning in myfind_cor_features(feats_df = glb_feats_df, obs_df =
## glbObsTrn, : Identified resXY.min.log1p as highly correlated with resXY.min## [1] "cor(resXY.mad, resXY.mad.root2)=0.9334"
## [1] "cor(outdoor.fctr, resXY.mad)=-0.0119"
## [1] "cor(outdoor.fctr, resXY.mad.root2)=-0.0114"## Warning in myfind_cor_features(feats_df = glb_feats_df, obs_df =
## glbObsTrn, : Identified resXY.mad.root2 as highly correlated with resXY.mad## [1] "cor(resX.mad.log1p, resX.mad.nexp)=-0.9321"
## [1] "cor(outdoor.fctr, resX.mad.log1p)=0.0220"
## [1] "cor(outdoor.fctr, resX.mad.nexp)=-0.0140"## Warning in myfind_cor_features(feats_df = glb_feats_df, obs_df =
## glbObsTrn, : Identified resX.mad.nexp as highly correlated with
## resX.mad.log1p## [1] "cor(nImgs.log1p, nImgs.root2)=0.9280"
## [1] "cor(outdoor.fctr, nImgs.log1p)=0.0473"
## [1] "cor(outdoor.fctr, nImgs.root2)=0.0140"## Warning in myfind_cor_features(feats_df = glb_feats_df, obs_df =
## glbObsTrn, : Identified nImgs.root2 as highly correlated with nImgs.log1p## [1] "cor(nImgs.cut.fctr, nImgs.log1p)=0.9109"
## [1] "cor(outdoor.fctr, nImgs.cut.fctr)=0.0586"
## [1] "cor(outdoor.fctr, nImgs.log1p)=0.0473"## Warning in myfind_cor_features(feats_df = glb_feats_df, obs_df
## = glbObsTrn, : Identified nImgs.log1p as highly correlated with
## nImgs.cut.fctr## [1] "cor(resXY.min, resY.min)=0.8858"
## [1] "cor(outdoor.fctr, resXY.min)=-0.0495"
## [1] "cor(outdoor.fctr, resY.min)=-0.0509"## Warning in myfind_cor_features(feats_df = glb_feats_df, obs_df =
## glbObsTrn, : Identified resXY.min as highly correlated with resY.min## cor.y exclude.as.feat cor.y.abs cor.high.X
## outdoor 1.000000000 1 1.000000000 <NA>
## liquor 0.100416198 1 0.100416198 <NA>
## nImgs.cut.fctr 0.058567974 0 0.058567974 <NA>
## nImgs.log1p 0.047250893 0 0.047250893 nImgs.cut.fctr
## reserve 0.038935338 1 0.038935338 <NA>
## resX.max.nexp 0.031543826 0 0.031543826 <NA>
## .pos 0.027497300 0 0.027497300 <NA>
## resX.mad.log1p 0.022032870 0 0.022032870 <NA>
## resX.mad.root2 0.021937317 0 0.021937317 resX.mad.log1p
## resX.mad 0.020537518 0 0.020537518 resX.mad.log1p
## resY.max.nexp 0.019090479 0 0.019090479 <NA>
## expensive 0.017228141 1 0.017228141 <NA>
## classy 0.015804825 1 0.015804825 <NA>
## resXY.mad.nexp 0.015437895 0 0.015437895 <NA>
## nImgs.root2 0.014028124 0 0.014028124 nImgs.log1p
## resY.mean.log1p 0.013625190 0 0.013625190 <NA>
## resY.mean.root2 0.013106506 0 0.013106506 resY.mean.log1p
## resY.mean 0.012599188 0 0.012599188 resY.mean.root2
## resY.mad.nexp 0.012190340 0 0.012190340 <NA>
## resY.max 0.011656712 0 0.011656712 <NA>
## resY.max.root2 0.011556200 0 0.011556200 resY.max
## resY.max.log1p 0.011451372 0 0.011451372 resY.max
## resY.mad 0.007630633 0 0.007630633 <NA>
## resXY.max.log1p 0.005240654 0 0.005240654 <NA>
## resXY.max.root2 0.004944889 0 0.004944889 <NA>
## resXY.max 0.004653277 0 0.004653277 <NA>
## resY.mad.root2 0.002557583 0 0.002557583 <NA>
## resY.mad.log1p -0.001526058 0 0.001526058 <NA>
## nImgs.nexp -0.003435316 0 0.003435316 <NA>
## resXY.mean.log1p -0.004867571 0 0.004867571 <NA>
## lunch -0.005308550 1 0.005308550 <NA>
## resXY.mean.root2 -0.007039955 0 0.007039955 <NA>
## .rnorm -0.008042720 0 0.008042720 <NA>
## resXY.mean -0.009002880 0 0.009002880 <NA>
## resXY.mad.root2 -0.011364822 0 0.011364822 resXY.mad
## resXY.mad -0.011946049 0 0.011946049 <NA>
## resX.mad.nexp -0.014000008 0 0.014000008 resX.mad.log1p
## resXY.mad.log1p -0.014055066 0 0.014055066 resXY.mad.nexp
## nImgs -0.014963676 0 0.014963676 <NA>
## resX.mean.log1p -0.015059015 0 0.015059015 resX.mean
## resX.mean.root2 -0.016434019 0 0.016434019 resX.mean
## resX.mean -0.017726551 0 0.017726551 <NA>
## resX.min.nexp -0.022391602 0 0.022391602 <NA>
## resX.mean.nexp -0.022433472 0 0.022433472 <NA>
## resY.mean.nexp -0.022433472 0 0.022433472 resX.mean.nexp
## resY.min.nexp -0.022433600 0 0.022433600 <NA>
## resX.min.log1p -0.030103276 0 0.030103276 resX.min
## resX.min.root2 -0.030339745 0 0.030339745 resX.min
## resX.min -0.031436275 0 0.031436275 <NA>
## resX.max -0.031543826 0 0.031543826 resX.max.nexp
## resX.max.log1p -0.031543826 0 0.031543826 resX.max
## resX.max.root2 -0.031543826 0 0.031543826 resX.max.nexp
## resXY.min.log1p -0.033756424 0 0.033756424 resXY.min
## dinner -0.039980159 1 0.039980159 <NA>
## resXY.min.root2 -0.041449898 0 0.041449898 resXY.min
## resY.min.log1p -0.043072548 0 0.043072548 resY.min
## resY.min.root2 -0.047387777 0 0.047387777 resY.min
## resXY.min -0.049458217 0 0.049458217 resY.min
## resY.min -0.050925308 0 0.050925308 <NA>
## table -0.055823041 1 0.055823041 <NA>
## kids -0.075895168 1 0.075895168 <NA>
## resXY.max.nexp NA 0 NA <NA>
## resXY.mean.nexp NA 0 NA <NA>
## resXY.min.nexp NA 0 NA <NA>
## freqRatio percentUnique zeroVar nzv is.cor.y.abs.low
## outdoor 1.006018 0.10 FALSE FALSE FALSE
## liquor 1.663116 0.10 FALSE FALSE FALSE
## nImgs.cut.fctr 1.007737 0.20 FALSE FALSE FALSE
## nImgs.log1p 1.033333 19.10 FALSE FALSE FALSE
## reserve 1.053388 0.10 FALSE FALSE FALSE
## resX.max.nexp 999.000000 0.10 FALSE TRUE FALSE
## .pos 1.000000 100.00 FALSE FALSE FALSE
## resX.mad.log1p 11.209677 7.55 FALSE FALSE FALSE
## resX.mad.root2 11.209677 7.55 FALSE FALSE FALSE
## resX.mad 11.209677 7.55 FALSE FALSE FALSE
## resY.max.nexp 1997.000000 0.20 FALSE TRUE FALSE
## expensive 2.656307 0.10 FALSE FALSE FALSE
## classy 2.496503 0.10 FALSE FALSE FALSE
## resXY.mad.nexp 4.841317 0.30 FALSE FALSE FALSE
## nImgs.root2 1.033333 19.10 FALSE FALSE FALSE
## resY.mean.log1p 1.666667 97.90 FALSE FALSE FALSE
## resY.mean.root2 1.666667 97.85 FALSE FALSE FALSE
## resY.mean 1.666667 98.15 FALSE FALSE FALSE
## resY.mad.nexp 5.354497 9.05 FALSE FALSE FALSE
## resY.max 1997.000000 0.20 FALSE TRUE FALSE
## resY.max.root2 1997.000000 0.20 FALSE TRUE FALSE
## resY.max.log1p 1997.000000 0.20 FALSE TRUE FALSE
## resY.mad 5.354497 9.05 FALSE FALSE TRUE
## resXY.max.log1p 6.272358 5.45 FALSE FALSE TRUE
## resXY.max.root2 6.272358 5.45 FALSE FALSE TRUE
## resXY.max 6.272358 5.45 FALSE FALSE TRUE
## resY.mad.root2 5.354497 9.05 FALSE FALSE TRUE
## resY.mad.log1p 5.354497 9.05 FALSE FALSE TRUE
## nImgs.nexp 1.193548 17.35 FALSE FALSE TRUE
## resXY.mean.log1p 4.000000 90.80 FALSE FALSE TRUE
## lunch 1.980626 0.10 FALSE FALSE TRUE
## resXY.mean.root2 6.000000 98.20 FALSE FALSE TRUE
## .rnorm 1.000000 100.00 FALSE FALSE FALSE
## resXY.mean 6.000000 98.55 FALSE FALSE FALSE
## resXY.mad.root2 9.568047 4.35 FALSE FALSE FALSE
## resXY.mad 9.568047 4.35 FALSE FALSE FALSE
## resX.mad.nexp 11.209677 7.55 FALSE FALSE FALSE
## resXY.mad.log1p 9.568047 4.35 FALSE FALSE FALSE
## nImgs 1.033333 19.10 FALSE FALSE FALSE
## resX.mean.log1p 2.000000 97.60 FALSE FALSE FALSE
## resX.mean.root2 2.000000 97.45 FALSE FALSE FALSE
## resX.mean 2.000000 97.75 FALSE FALSE FALSE
## resX.min.nexp 6.000000 11.45 FALSE FALSE FALSE
## resX.mean.nexp 2.000000 97.75 FALSE FALSE FALSE
## resY.mean.nexp 1.666667 98.15 FALSE FALSE FALSE
## resY.min.nexp 9.824561 13.85 FALSE FALSE FALSE
## resX.min.log1p 6.000000 11.45 FALSE FALSE FALSE
## resX.min.root2 6.000000 11.45 FALSE FALSE FALSE
## resX.min 6.000000 11.45 FALSE FALSE FALSE
## resX.max 999.000000 0.10 FALSE TRUE FALSE
## resX.max.log1p 999.000000 0.10 FALSE TRUE FALSE
## resX.max.root2 999.000000 0.10 FALSE TRUE FALSE
## resXY.min.log1p 9.745455 37.65 FALSE FALSE FALSE
## dinner 1.014099 0.10 FALSE FALSE FALSE
## resXY.min.root2 9.745455 37.65 FALSE FALSE FALSE
## resY.min.log1p 9.824561 13.85 FALSE FALSE FALSE
## resY.min.root2 9.824561 13.85 FALSE FALSE FALSE
## resXY.min 9.745455 37.65 FALSE FALSE FALSE
## resY.min 9.824561 13.85 FALSE FALSE FALSE
## table 2.125000 0.10 FALSE FALSE FALSE
## kids 1.624672 0.10 FALSE FALSE FALSE
## resXY.max.nexp 0.000000 0.05 TRUE TRUE NA
## resXY.mean.nexp 0.000000 0.05 TRUE TRUE NA
## resXY.min.nexp 0.000000 0.05 TRUE TRUE NA## Warning in myplot_scatter(plt_feats_df, "percentUnique", "freqRatio",
## colorcol_name = "nzv", : converting nzv to class:factor## Warning: Removed 16 rows containing missing values (geom_point).
## Warning: Removed 16 rows containing missing values (geom_point).
## Warning: Removed 16 rows containing missing values (geom_point).
## cor.y exclude.as.feat cor.y.abs cor.high.X
## resX.max.nexp 0.03154383 0 0.03154383 <NA>
## resY.max.nexp 0.01909048 0 0.01909048 <NA>
## resY.max 0.01165671 0 0.01165671 <NA>
## resY.max.root2 0.01155620 0 0.01155620 resY.max
## resY.max.log1p 0.01145137 0 0.01145137 resY.max
## resX.max -0.03154383 0 0.03154383 resX.max.nexp
## resX.max.log1p -0.03154383 0 0.03154383 resX.max
## resX.max.root2 -0.03154383 0 0.03154383 resX.max.nexp
## resXY.max.nexp NA 0 NA <NA>
## resXY.mean.nexp NA 0 NA <NA>
## resXY.min.nexp NA 0 NA <NA>
## freqRatio percentUnique zeroVar nzv is.cor.y.abs.low
## resX.max.nexp 999 0.10 FALSE TRUE FALSE
## resY.max.nexp 1997 0.20 FALSE TRUE FALSE
## resY.max 1997 0.20 FALSE TRUE FALSE
## resY.max.root2 1997 0.20 FALSE TRUE FALSE
## resY.max.log1p 1997 0.20 FALSE TRUE FALSE
## resX.max 999 0.10 FALSE TRUE FALSE
## resX.max.log1p 999 0.10 FALSE TRUE FALSE
## resX.max.root2 999 0.10 FALSE TRUE FALSE
## resXY.max.nexp 0 0.05 TRUE TRUE NA
## resXY.mean.nexp 0 0.05 TRUE TRUE NA
## resXY.min.nexp 0 0.05 TRUE TRUE NA
## [1] "numeric data missing in glbObsAll: "
## lunch dinner reserve outdoor expensive
## 10000 10000 10000 10000 10000
## liquor table classy kids outdoor.fctr
## 10000 10000 10000 10000 10000
## [1] "numeric data w/ 0s in glbObsAll: "
## nImgs.nexp resX.mad resX.mad.log1p resX.mad.root2
## 228 9353 9353 9353
## resY.mad resY.mad.log1p resY.mad.root2 resXY.mad
## 5442 5442 5442 10915
## resXY.min.nexp resXY.max.nexp resXY.mean.nexp resXY.mad.log1p
## 12000 12000 12000 10915
## resXY.mad.root2 resXY.mad.nexp lunch
## 10915 850 671
## [1] "numeric data w/ Infs in glbObsAll: "
## named integer(0)
## [1] "numeric data w/ NaNs in glbObsAll: "
## named integer(0)
## [1] "string data missing in glbObsAll: "
## business_id labels resXLst resYLst .lcn
## 0 NA 0 0 10000## [1] "glb_feats_df:"## [1] 64 12## id exclude.as.feat rsp_var
## outdoor.fctr outdoor.fctr TRUE TRUE## id cor.y exclude.as.feat cor.y.abs cor.high.X
## outdoor outdoor 1 TRUE 1 <NA>
## outdoor.fctr outdoor.fctr NA TRUE NA <NA>
## freqRatio percentUnique zeroVar nzv is.cor.y.abs.low
## outdoor 1.006018 0.1 FALSE FALSE FALSE
## outdoor.fctr NA NA NA NA NA
## interaction.feat shapiro.test.p.value rsp_var_raw id_var
## outdoor NA NA TRUE NA
## outdoor.fctr NA NA NA NA
## rsp_var
## outdoor NA
## outdoor.fctr TRUE## [1] "glb_feats_df vs. glbObsAll: "## character(0)## [1] "glbObsAll vs. glb_feats_df: "## character(0)## label step_major step_minor label_minor bgn end
## 15 select.features 7 0 0 129.924 132.968
## 16 fit.models 8 0 0 132.969 NA
## elapsed
## 15 3.045
## 16 NAStep 8.0: fit models
fit.models_0_chunk_df <- myadd_chunk(NULL, "fit.models_0_bgn", label.minor = "setup")## label step_major step_minor label_minor bgn end elapsed
## 1 fit.models_0_bgn 1 0 setup 133.588 NA NA# load(paste0(glbOut$pfx, "dsk.RData"))
get_model_sel_frmla <- function() {
model_evl_terms <- c(NULL)
# min.aic.fit might not be avl
lclMdlEvlCriteria <-
glbMdlMetricsEval[glbMdlMetricsEval %in% names(glb_models_df)]
for (metric in lclMdlEvlCriteria)
model_evl_terms <- c(model_evl_terms,
ifelse(length(grep("max", metric)) > 0, "-", "+"), metric)
if (glb_is_classification && glb_is_binomial)
model_evl_terms <- c(model_evl_terms, "-", "opt.prob.threshold.OOB")
model_sel_frmla <- as.formula(paste(c("~ ", model_evl_terms), collapse = " "))
return(model_sel_frmla)
}
get_dsp_models_df <- function() {
dsp_models_cols <- c("id",
glbMdlMetricsEval[glbMdlMetricsEval %in% names(glb_models_df)],
grep("opt.", names(glb_models_df), fixed = TRUE, value = TRUE))
dsp_models_df <-
#orderBy(get_model_sel_frmla(), glb_models_df)[, c("id", glbMdlMetricsEval)]
orderBy(get_model_sel_frmla(), glb_models_df)[, dsp_models_cols]
nCvMdl <- sapply(glb_models_lst, function(mdl) nrow(mdl$results))
nParams <- sapply(glb_models_lst, function(mdl) ifelse(mdl$method == "custom", 0,
nrow(subset(modelLookup(mdl$method), parameter != "parameter"))))
# nCvMdl <- nCvMdl[names(nCvMdl) != "avNNet"]
# nParams <- nParams[names(nParams) != "avNNet"]
if (length(cvMdlProblems <- nCvMdl[nCvMdl <= nParams]) > 0) {
print("Cross Validation issues:")
warning("Cross Validation issues:")
print(cvMdlProblems)
}
pltMdls <- setdiff(names(nCvMdl), names(cvMdlProblems))
pltMdls <- setdiff(pltMdls, names(nParams[nParams == 0]))
# length(pltMdls) == 21
png(paste0(glbOut$pfx, "bestTune.png"), width = 480 * 2, height = 480 * 4)
grid.newpage()
pushViewport(viewport(layout = grid.layout(ceiling(length(pltMdls) / 2.0), 2)))
pltIx <- 1
for (mdlId in pltMdls) {
print(ggplot(glb_models_lst[[mdlId]], highBestTune = TRUE) + labs(title = mdlId),
vp = viewport(layout.pos.row = ceiling(pltIx / 2.0),
layout.pos.col = ((pltIx - 1) %% 2) + 1))
pltIx <- pltIx + 1
}
dev.off()
if (all(row.names(dsp_models_df) != dsp_models_df$id))
row.names(dsp_models_df) <- dsp_models_df$id
return(dsp_models_df)
}
#get_dsp_models_df()
if (glb_is_classification && glb_is_binomial &&
(length(unique(glbObsFit[, glb_rsp_var])) < 2))
stop("glbObsFit$", glb_rsp_var, ": contains less than 2 unique values: ",
paste0(unique(glbObsFit[, glb_rsp_var]), collapse=", "))
max_cor_y_x_vars <- orderBy(~ -cor.y.abs,
subset(glb_feats_df, (exclude.as.feat == 0) & !nzv & !is.cor.y.abs.low &
is.na(cor.high.X)))[1:2, "id"]
max_cor_y_x_vars <- max_cor_y_x_vars[!is.na(max_cor_y_x_vars)]
if (length(max_cor_y_x_vars) < 2)
max_cor_y_x_vars <- union(max_cor_y_x_vars, ".pos")
if (!is.null(glb_Baseline_mdl_var)) {
if ((max_cor_y_x_vars[1] != glb_Baseline_mdl_var) &
(glb_feats_df[glb_feats_df$id == max_cor_y_x_vars[1], "cor.y.abs"] >
glb_feats_df[glb_feats_df$id == glb_Baseline_mdl_var, "cor.y.abs"]))
stop(max_cor_y_x_vars[1], " has a higher correlation with ", glb_rsp_var,
" than the Baseline var: ", glb_Baseline_mdl_var)
}
glb_model_type <- ifelse(glb_is_regression, "regression", "classification")
# Model specs
c("id.prefix", "method", "type",
# trainControl params
"preProc.method", "cv.n.folds", "cv.n.repeats", "summary.fn",
# train params
"metric", "metric.maximize", "tune.df")## [1] "id.prefix" "method" "type"
## [4] "preProc.method" "cv.n.folds" "cv.n.repeats"
## [7] "summary.fn" "metric" "metric.maximize"
## [10] "tune.df"# Baseline
if (!is.null(glb_Baseline_mdl_var)) {
fit.models_0_chunk_df <- myadd_chunk(fit.models_0_chunk_df,
paste0("fit.models_0_", "Baseline"), major.inc = FALSE,
label.minor = "mybaseln_classfr")
ret_lst <- myfit_mdl(mdl_id="Baseline",
model_method="mybaseln_classfr",
indep_vars_vctr=glb_Baseline_mdl_var,
rsp_var=glb_rsp_var,
fit_df=glbObsFit, OOB_df=glbObsOOB)
}
# Most Frequent Outcome "MFO" model: mean(y) for regression
# Not using caret's nullModel since model stats not avl
# Cannot use rpart for multinomial classification since it predicts non-MFO
if (glb_is_classification) {
fit.models_0_chunk_df <- myadd_chunk(fit.models_0_chunk_df,
paste0("fit.models_0_", "MFO"), major.inc = FALSE,
label.minor = "myMFO_classfr")
ret_lst <- myfit_mdl(mdl_specs_lst = myinit_mdl_specs_lst(mdl_specs_lst = list(
id.prefix = "MFO", type = glb_model_type, trainControl.method = "none",
train.method = ifelse(glb_is_regression, "lm", "myMFO_classfr"))),
indep_vars = ".rnorm", rsp_var = glb_rsp_var,
fit_df = glbObsFit, OOB_df = glbObsOOB)
# "random" model - only for classification;
# none needed for regression since it is same as MFO
fit.models_0_chunk_df <- myadd_chunk(fit.models_0_chunk_df,
paste0("fit.models_0_", "Random"), major.inc = FALSE,
label.minor = "myrandom_classfr")
#stop(here"); glb2Sav(); all.equal(glb_models_df, sav_models_df)
ret_lst <- myfit_mdl(mdl_specs_lst = myinit_mdl_specs_lst(mdl_specs_lst = list(
id.prefix = "Random", type = glb_model_type, trainControl.method = "none",
train.method = "myrandom_classfr")),
indep_vars = ".rnorm", rsp_var = glb_rsp_var,
fit_df = glbObsFit, OOB_df = glbObsOOB)
}## label step_major step_minor label_minor bgn end
## 1 fit.models_0_bgn 1 0 setup 133.588 133.622
## 2 fit.models_0_MFO 1 1 myMFO_classfr 133.623 NA
## elapsed
## 1 0.034
## 2 NA
## [1] "myfit_mdl: enter: 0.000000 secs"
## [1] "fitting model: MFO###myMFO_classfr"
## [1] " indep_vars: .rnorm"
## [1] "myfit_mdl: setup complete: 0.412000 secs"
## Fitting parameter = none on full training set
## [1] "in MFO.Classifier$fit"
## [1] "unique.vals:"
## [1] N Y
## Levels: N Y
## [1] "unique.prob:"
## y
## Y N
## 0.500998 0.499002
## [1] "MFO.val:"
## [1] "Y"
## [1] "myfit_mdl: train complete: 0.841000 secs"
## Length Class Mode
## unique.vals 2 factor numeric
## unique.prob 2 -none- numeric
## MFO.val 1 -none- character
## x.names 1 -none- character
## xNames 1 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 2 -none- character
## [1] "myfit_mdl: train diagnostics complete: 0.843000 secs"## Loading required namespace: pROC## [1] "entr MFO.Classifier$predict"
## [1] "exit MFO.Classifier$predict"## Loading required package: ROCR## Loading required package: gplots##
## Attaching package: 'gplots'## The following object is masked from 'package:stats':
##
## lowess## [1] "in MFO.Classifier$prob"
## N Y
## 1 0.500998 0.499002
## 2 0.500998 0.499002
## 3 0.500998 0.499002
## 4 0.500998 0.499002
## 5 0.500998 0.499002
## 6 0.500998 0.499002
## Prediction
## Reference N Y
## N 0 500
## Y 0 502
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 5.009980e-01 0.000000e+00 4.695761e-01 5.324140e-01 5.009980e-01
## AccuracyPValue McnemarPValue
## 5.126083e-01 2.586405e-110
## [1] "entr MFO.Classifier$predict"
## [1] "exit MFO.Classifier$predict"
## [1] "in MFO.Classifier$prob"
## N Y
## 1 0.500998 0.499002
## 2 0.500998 0.499002
## 3 0.500998 0.499002
## 4 0.500998 0.499002
## 5 0.500998 0.499002
## 6 0.500998 0.499002
## Prediction
## Reference N Y
## N 0 497
## Y 0 501
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 5.020040e-01 0.000000e+00 4.705156e-01 5.334806e-01 5.020040e-01
## AccuracyPValue McnemarPValue
## 5.126421e-01 1.162632e-109
## [1] "myfit_mdl: predict complete: 3.129000 secs"
## id feats max.nTuningRuns min.elapsedtime.everything
## 1 MFO###myMFO_classfr .rnorm 0 0.422
## min.elapsedtime.final max.AUCpROC.fit max.Sens.fit max.Spec.fit
## 1 0.002 0.5 0 1
## max.AUCROCR.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.5 0.4 0.6675532 0.500998
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit
## 1 0.4695761 0.532414 0
## max.AUCpROC.OOB max.Sens.OOB max.Spec.OOB max.AUCROCR.OOB
## 1 0.5 0 1 0.5
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.4 0.6684456 0.502004
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.4705156 0.5334806 0
## [1] "myfit_mdl: exit: 3.139000 secs"
## label step_major step_minor label_minor bgn
## 2 fit.models_0_MFO 1 1 myMFO_classfr 133.623
## 3 fit.models_0_Random 1 2 myrandom_classfr 136.767
## end elapsed
## 2 136.767 3.144
## 3 NA NA
## [1] "myfit_mdl: enter: 0.000000 secs"
## [1] "fitting model: Random###myrandom_classfr"
## [1] " indep_vars: .rnorm"
## [1] "myfit_mdl: setup complete: 0.427000 secs"
## Fitting parameter = none on full training set
## [1] "myfit_mdl: train complete: 0.774000 secs"
## Length Class Mode
## unique.vals 2 factor numeric
## unique.prob 2 table numeric
## xNames 1 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 2 -none- character
## [1] "myfit_mdl: train diagnostics complete: 0.775000 secs"
## [1] "in Random.Classifier$prob"
## Prediction
## Reference N Y
## N 0 500
## Y 0 502
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 5.009980e-01 0.000000e+00 4.695761e-01 5.324140e-01 5.009980e-01
## AccuracyPValue McnemarPValue
## 5.126083e-01 2.586405e-110
## [1] "in Random.Classifier$prob"
## Prediction
## Reference N Y
## N 0 497
## Y 0 501
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 5.020040e-01 0.000000e+00 4.705156e-01 5.334806e-01 5.020040e-01
## AccuracyPValue McnemarPValue
## 5.126421e-01 1.162632e-109
## [1] "myfit_mdl: predict complete: 3.343000 secs"
## id feats max.nTuningRuns
## 1 Random###myrandom_classfr .rnorm 0
## min.elapsedtime.everything min.elapsedtime.final max.AUCpROC.fit
## 1 0.343 0.002 0.4980239
## max.Sens.fit max.Spec.fit max.AUCROCR.fit opt.prob.threshold.fit
## 1 0.508 0.4880478 0.487012 0.4
## max.f.score.fit max.Accuracy.fit max.AccuracyLower.fit
## 1 0.6675532 0.500998 0.4695761
## max.AccuracyUpper.fit max.Kappa.fit max.AUCpROC.OOB max.Sens.OOB
## 1 0.532414 0 0.5059679 0.4949698
## max.Spec.OOB max.AUCROCR.OOB opt.prob.threshold.OOB max.f.score.OOB
## 1 0.5169661 0.4969618 0.4 0.6684456
## max.Accuracy.OOB max.AccuracyLower.OOB max.AccuracyUpper.OOB
## 1 0.502004 0.4705156 0.5334806
## max.Kappa.OOB
## 1 0
## [1] "myfit_mdl: exit: 3.356000 secs"# Max.cor.Y
# Check impact of cv
# rpart is not a good candidate since caret does not optimize cp (only tuning parameter of rpart) well
fit.models_0_chunk_df <- myadd_chunk(fit.models_0_chunk_df,
paste0("fit.models_0_", "Max.cor.Y.rcv.*X*"), major.inc = FALSE,
label.minor = "glmnet")## label step_major step_minor label_minor
## 3 fit.models_0_Random 1 2 myrandom_classfr
## 4 fit.models_0_Max.cor.Y.rcv.*X* 1 3 glmnet
## bgn end elapsed
## 3 136.767 140.136 3.369
## 4 140.137 NA NAret_lst <- myfit_mdl(mdl_specs_lst=myinit_mdl_specs_lst(mdl_specs_lst=list(
id.prefix="Max.cor.Y.rcv.1X1", type=glb_model_type, trainControl.method="none",
train.method="glmnet")),
indep_vars=max_cor_y_x_vars, rsp_var=glb_rsp_var,
fit_df = glbObsFit, OOB_df = glbObsOOB)## [1] "myfit_mdl: enter: 0.000000 secs"
## [1] "fitting model: Max.cor.Y.rcv.1X1###glmnet"
## [1] " indep_vars: nImgs.cut.fctr,resY.min"
## [1] "myfit_mdl: setup complete: 0.674000 secs"## Loading required package: glmnet## Loading required package: Matrix## Loaded glmnet 2.0-2## Fitting alpha = 0.1, lambda = 0.000488 on full training set
## [1] "myfit_mdl: train complete: 1.460000 secs"
## Length Class Mode
## a0 41 -none- numeric
## beta 164 dgCMatrix S4
## df 41 -none- numeric
## dim 2 -none- numeric
## lambda 41 -none- numeric
## dev.ratio 41 -none- numeric
## nulldev 1 -none- numeric
## npasses 1 -none- numeric
## jerr 1 -none- numeric
## offset 1 -none- logical
## classnames 2 -none- character
## call 5 -none- call
## nobs 1 -none- numeric
## lambdaOpt 1 -none- numeric
## xNames 4 -none- character
## problemType 1 -none- character
## tuneValue 2 data.frame list
## obsLevels 2 -none- character
## [1] "min lambda > lambdaOpt:"
## (Intercept) nImgs.cut.fctr(32,60]
## -0.2246350 0.2599168
## nImgs.cut.fctr(60,120] nImgs.cut.fctr(120,3e+03]
## 0.2554564 0.3976048
## [1] "max lambda < lambdaOpt:"
## [1] "Feats mismatch between coefs_left & rght:"
## [1] "(Intercept)" "nImgs.cut.fctr(32,60]"
## [3] "nImgs.cut.fctr(60,120]" "nImgs.cut.fctr(120,3e+03]"
## [5] "resY.min"
## [1] "myfit_mdl: train diagnostics complete: 1.580000 secs"
## Prediction
## Reference N Y
## N 0 500
## Y 0 502
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 5.009980e-01 0.000000e+00 4.695761e-01 5.324140e-01 5.009980e-01
## AccuracyPValue McnemarPValue
## 5.126083e-01 2.586405e-110
## Prediction
## Reference N Y
## N 0 497
## Y 0 501
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 5.020040e-01 0.000000e+00 4.705156e-01 5.334806e-01 5.020040e-01
## AccuracyPValue McnemarPValue
## 5.126421e-01 1.162632e-109
## [1] "myfit_mdl: predict complete: 3.557000 secs"
## id feats max.nTuningRuns
## 1 Max.cor.Y.rcv.1X1###glmnet nImgs.cut.fctr,resY.min 0
## min.elapsedtime.everything min.elapsedtime.final max.AUCpROC.fit
## 1 0.78 0.017 0.5304143
## max.Sens.fit max.Spec.fit max.AUCROCR.fit opt.prob.threshold.fit
## 1 0.268 0.7928287 0.539247 0.4
## max.f.score.fit max.Accuracy.fit max.AccuracyLower.fit
## 1 0.6675532 0.500998 0.4695761
## max.AccuracyUpper.fit max.Kappa.fit max.AUCpROC.OOB max.Sens.OOB
## 1 0.532414 0 0.5300104 0.2676056
## max.Spec.OOB max.AUCROCR.OOB opt.prob.threshold.OOB max.f.score.OOB
## 1 0.7924152 0.5290506 0.4 0.6684456
## max.Accuracy.OOB max.AccuracyLower.OOB max.AccuracyUpper.OOB
## 1 0.502004 0.4705156 0.5334806
## max.Kappa.OOB
## 1 0
## [1] "myfit_mdl: exit: 3.570000 secs"if (glbMdlCheckRcv) {
# rcv_n_folds == 1 & rcv_n_repeats > 1 crashes
for (rcv_n_folds in seq(3, glb_rcv_n_folds + 2, 2))
for (rcv_n_repeats in seq(1, glb_rcv_n_repeats + 2, 2)) {
# Experiment specific code to avoid caret crash
# lcl_tune_models_df <- rbind(data.frame()
# ,data.frame(method = "glmnet", parameter = "alpha",
# vals = "0.100 0.325 0.550 0.775 1.000")
# ,data.frame(method = "glmnet", parameter = "lambda",
# vals = "9.342e-02")
# )
ret_lst <- myfit_mdl(mdl_specs_lst = myinit_mdl_specs_lst(mdl_specs_lst =
list(
id.prefix = paste0("Max.cor.Y.rcv.", rcv_n_folds, "X", rcv_n_repeats),
type = glb_model_type,
# tune.df = lcl_tune_models_df,
trainControl.method = "repeatedcv",
trainControl.number = rcv_n_folds,
trainControl.repeats = rcv_n_repeats,
trainControl.classProbs = glb_is_classification,
trainControl.summaryFunction = glbMdlMetricSummaryFn,
train.method = "glmnet", train.metric = glbMdlMetricSummary,
train.maximize = glbMdlMetricMaximize)),
indep_vars = max_cor_y_x_vars, rsp_var = glb_rsp_var,
fit_df = glbObsFit, OOB_df = glbObsOOB)
}
# Add parallel coordinates graph of glb_models_df[, glbMdlMetricsEval] to evaluate cv parameters
tmp_models_cols <- c("id", "max.nTuningRuns",
glbMdlMetricsEval[glbMdlMetricsEval %in% names(glb_models_df)],
grep("opt.", names(glb_models_df), fixed = TRUE, value = TRUE))
print(myplot_parcoord(obs_df = subset(glb_models_df,
grepl("Max.cor.Y.rcv.", id, fixed = TRUE),
select = -feats)[, tmp_models_cols],
id_var = "id"))
}
# Useful for stacking decisions
# fit.models_0_chunk_df <- myadd_chunk(fit.models_0_chunk_df,
# paste0("fit.models_0_", "Max.cor.Y[rcv.1X1.cp.0|]"), major.inc = FALSE,
# label.minor = "rpart")
#
# ret_lst <- myfit_mdl(mdl_specs_lst = myinit_mdl_specs_lst(mdl_specs_lst = list(
# id.prefix = "Max.cor.Y.rcv.1X1.cp.0", type = glb_model_type, trainControl.method = "none",
# train.method = "rpart",
# tune.df=data.frame(method="rpart", parameter="cp", min=0.0, max=0.0, by=0.1))),
# indep_vars=max_cor_y_x_vars, rsp_var=glb_rsp_var,
# fit_df=glbObsFit, OOB_df=glbObsOOB)
#stop(here"); glb2Sav(); all.equal(glb_models_df, sav_models_df)
# if (glb_is_regression || glb_is_binomial) # For multinomials this model will be run next by default
ret_lst <- myfit_mdl(mdl_specs_lst = myinit_mdl_specs_lst(mdl_specs_lst = list(
id.prefix = "Max.cor.Y",
type = glb_model_type, trainControl.method = "repeatedcv",
trainControl.number = glb_rcv_n_folds,
trainControl.repeats = glb_rcv_n_repeats,
trainControl.classProbs = glb_is_classification,
trainControl.summaryFunction = glbMdlMetricSummaryFn,
trainControl.allowParallel = glbMdlAllowParallel,
train.metric = glbMdlMetricSummary,
train.maximize = glbMdlMetricMaximize,
train.method = "rpart")),
indep_vars = max_cor_y_x_vars, rsp_var = glb_rsp_var,
fit_df = glbObsFit, OOB_df = glbObsOOB)## [1] "myfit_mdl: enter: 0.000000 secs"
## [1] "fitting model: Max.cor.Y##rcv#rpart"
## [1] " indep_vars: nImgs.cut.fctr,resY.min"
## [1] "myfit_mdl: setup complete: 0.669000 secs"## Loading required package: rpart## Aggregating results
## Selecting tuning parameters
## Fitting cp = 0.012 on full training set
## [1] "myfit_mdl: train complete: 2.365000 secs"## Warning in myfit_mdl(mdl_specs_lst = myinit_mdl_specs_lst(mdl_specs_lst
## = list(id.prefix = "Max.cor.Y", : model's bestTune found at an extreme of
## tuneGrid for parameter: cp## Loading required package: rpart.plot
## Call:
## rpart(formula = .outcome ~ ., control = list(minsplit = 20, minbucket = 7,
## cp = 0, maxcompete = 4, maxsurrogate = 5, usesurrogate = 2,
## surrogatestyle = 0, maxdepth = 30, xval = 0))
## n= 1002
##
## CP nsplit rel error
## 1 0.070 0 1.000
## 2 0.044 1 0.930
## 3 0.017 2 0.886
## 4 0.014 4 0.852
## 5 0.012 5 0.838
##
## Variable importance
## resY.min nImgs.cut.fctr(60,120]
## 89 7
## nImgs.cut.fctr(120,3e+03] nImgs.cut.fctr(32,60]
## 3 1
##
## Node number 1: 1002 observations, complexity param=0.07
## predicted class=Y expected loss=0.499002 P(node) =1
## class counts: 500 502
## probabilities: 0.499 0.501
## left son=2 (509 obs) right son=3 (493 obs)
## Primary splits:
## resY.min < 261.5 to the right, improve=2.58978200, (0 missing)
## nImgs.cut.fctr(120,3e+03] < 0.5 to the left, improve=1.19439200, (0 missing)
## nImgs.cut.fctr(32,60] < 0.5 to the left, improve=0.07380192, (0 missing)
## nImgs.cut.fctr(60,120] < 0.5 to the left, improve=0.05268239, (0 missing)
## Surrogate splits:
## nImgs.cut.fctr(120,3e+03] < 0.5 to the left, agree=0.589, adj=0.164, (0 split)
## nImgs.cut.fctr(32,60] < 0.5 to the right, agree=0.538, adj=0.061, (0 split)
## nImgs.cut.fctr(60,120] < 0.5 to the left, agree=0.525, adj=0.034, (0 split)
##
## Node number 2: 509 observations, complexity param=0.017
## predicted class=N expected loss=0.4656189 P(node) =0.507984
## class counts: 272 237
## probabilities: 0.534 0.466
## left son=4 (389 obs) right son=5 (120 obs)
## Primary splits:
## nImgs.cut.fctr(60,120] < 0.5 to the left, improve=1.107328000, (0 missing)
## resY.min < 372.5 to the right, improve=1.096090000, (0 missing)
## nImgs.cut.fctr(32,60] < 0.5 to the left, improve=0.378170100, (0 missing)
## nImgs.cut.fctr(120,3e+03] < 0.5 to the left, improve=0.009365927, (0 missing)
##
## Node number 3: 493 observations, complexity param=0.044
## predicted class=Y expected loss=0.4624746 P(node) =0.492016
## class counts: 228 265
## probabilities: 0.462 0.538
## left son=6 (130 obs) right son=7 (363 obs)
## Primary splits:
## resY.min < 136 to the left, improve=5.26786600, (0 missing)
## nImgs.cut.fctr(120,3e+03] < 0.5 to the left, improve=0.84070860, (0 missing)
## nImgs.cut.fctr(60,120] < 0.5 to the right, improve=0.64330140, (0 missing)
## nImgs.cut.fctr(32,60] < 0.5 to the right, improve=0.00280855, (0 missing)
##
## Node number 4: 389 observations
## predicted class=N expected loss=0.4473008 P(node) =0.3882236
## class counts: 215 174
## probabilities: 0.553 0.447
##
## Node number 5: 120 observations, complexity param=0.017
## predicted class=Y expected loss=0.475 P(node) =0.1197605
## class counts: 57 63
## probabilities: 0.475 0.525
## left son=10 (23 obs) right son=11 (97 obs)
## Primary splits:
## resY.min < 280.5 to the left, improve=3.970126, (0 missing)
##
## Node number 6: 130 observations, complexity param=0.014
## predicted class=N expected loss=0.4153846 P(node) =0.1297405
## class counts: 76 54
## probabilities: 0.585 0.415
## left son=12 (93 obs) right son=13 (37 obs)
## Primary splits:
## resY.min < 94.5 to the right, improve=3.32212900, (0 missing)
## nImgs.cut.fctr(32,60] < 0.5 to the right, improve=0.13810200, (0 missing)
## nImgs.cut.fctr(60,120] < 0.5 to the right, improve=0.06990362, (0 missing)
## nImgs.cut.fctr(120,3e+03] < 0.5 to the left, improve=0.02157842, (0 missing)
##
## Node number 7: 363 observations
## predicted class=Y expected loss=0.4187328 P(node) =0.3622754
## class counts: 152 211
## probabilities: 0.419 0.581
##
## Node number 10: 23 observations
## predicted class=N expected loss=0.2608696 P(node) =0.02295409
## class counts: 17 6
## probabilities: 0.739 0.261
##
## Node number 11: 97 observations
## predicted class=Y expected loss=0.4123711 P(node) =0.09680639
## class counts: 40 57
## probabilities: 0.412 0.588
##
## Node number 12: 93 observations
## predicted class=N expected loss=0.344086 P(node) =0.09281437
## class counts: 61 32
## probabilities: 0.656 0.344
##
## Node number 13: 37 observations
## predicted class=Y expected loss=0.4054054 P(node) =0.03692615
## class counts: 15 22
## probabilities: 0.405 0.595
##
## n= 1002
##
## node), split, n, loss, yval, (yprob)
## * denotes terminal node
##
## 1) root 1002 500 Y (0.4990020 0.5009980)
## 2) resY.min>=261.5 509 237 N (0.5343811 0.4656189)
## 4) nImgs.cut.fctr(60,120]< 0.5 389 174 N (0.5526992 0.4473008) *
## 5) nImgs.cut.fctr(60,120]>=0.5 120 57 Y (0.4750000 0.5250000)
## 10) resY.min< 280.5 23 6 N (0.7391304 0.2608696) *
## 11) resY.min>=280.5 97 40 Y (0.4123711 0.5876289) *
## 3) resY.min< 261.5 493 228 Y (0.4624746 0.5375254)
## 6) resY.min< 136 130 54 N (0.5846154 0.4153846)
## 12) resY.min>=94.5 93 32 N (0.6559140 0.3440860) *
## 13) resY.min< 94.5 37 15 Y (0.4054054 0.5945946) *
## 7) resY.min>=136 363 152 Y (0.4187328 0.5812672) *
## [1] "myfit_mdl: train diagnostics complete: 3.404000 secs"
## Prediction
## Reference N Y
## N 17 483
## Y 6 496
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 5.119760e-01 2.208979e-02 4.805306e-01 5.433509e-01 5.009980e-01
## AccuracyPValue McnemarPValue
## 2.535456e-01 8.991577e-103
## Prediction
## Reference N Y
## N 0 497
## Y 0 501
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 5.020040e-01 0.000000e+00 4.705156e-01 5.334806e-01 5.020040e-01
## AccuracyPValue McnemarPValue
## 5.126421e-01 1.162632e-109
## [1] "myfit_mdl: predict complete: 5.515000 secs"
## id feats max.nTuningRuns
## 1 Max.cor.Y##rcv#rpart nImgs.cut.fctr,resY.min 5
## min.elapsedtime.everything min.elapsedtime.final max.AUCpROC.fit
## 1 1.69 0.016 0.5818446
## max.Sens.fit max.Spec.fit max.AUCROCR.fit opt.prob.threshold.fit
## 1 0.586 0.5776892 0.5938127 0.3
## max.f.score.fit max.Accuracy.fit max.AccuracyLower.fit
## 1 0.6698177 0.5255967 0.4805306
## max.AccuracyUpper.fit max.Kappa.fit max.AUCpROC.OOB max.Sens.OOB
## 1 0.5433509 0.05146235 0.5039318 0.4849095
## max.Spec.OOB max.AUCROCR.OOB opt.prob.threshold.OOB max.f.score.OOB
## 1 0.5229541 0.4985602 0.2 0.6684456
## max.Accuracy.OOB max.AccuracyLower.OOB max.AccuracyUpper.OOB
## 1 0.502004 0.4705156 0.5334806
## max.Kappa.OOB max.AccuracySD.fit max.KappaSD.fit
## 1 0 0.02756072 0.05485113
## [1] "myfit_mdl: exit: 5.530000 secs"if ((length(glbFeatsDateTime) > 0) &&
(sum(grepl(paste(names(glbFeatsDateTime), "\\.day\\.minutes\\.poly\\.", sep = ""),
names(glbObsAll))) > 0)) {
fit.models_0_chunk_df <- myadd_chunk(fit.models_0_chunk_df,
paste0("fit.models_0_", "Max.cor.Y.Time.Poly"), major.inc = FALSE,
label.minor = "glmnet")
indepVars <- c(max_cor_y_x_vars,
grep(paste(names(glbFeatsDateTime), "\\.day\\.minutes\\.poly\\.", sep = ""),
names(glbObsAll), value = TRUE))
indepVars <- myadjust_interaction_feats(indepVars)
ret_lst <- myfit_mdl(mdl_specs_lst = myinit_mdl_specs_lst(mdl_specs_lst = list(
id.prefix = "Max.cor.Y.Time.Poly",
type = glb_model_type, trainControl.method = "repeatedcv",
trainControl.number = glb_rcv_n_folds, trainControl.repeats = glb_rcv_n_repeats,
trainControl.classProbs = glb_is_classification,
trainControl.summaryFunction = glbMdlMetricSummaryFn,
train.metric = glbMdlMetricSummary,
train.maximize = glbMdlMetricMaximize,
train.method = "glmnet")),
indep_vars = indepVars,
rsp_var = glb_rsp_var,
fit_df = glbObsFit, OOB_df = glbObsOOB)
}
if ((length(glbFeatsDateTime) > 0) &&
(sum(grepl(paste(names(glbFeatsDateTime), "\\.last[[:digit:]]", sep = ""),
names(glbObsAll))) > 0)) {
fit.models_0_chunk_df <- myadd_chunk(fit.models_0_chunk_df,
paste0("fit.models_0_", "Max.cor.Y.Time.Lag"), major.inc = FALSE,
label.minor = "glmnet")
indepVars <- c(max_cor_y_x_vars,
grep(paste(names(glbFeatsDateTime), "\\.last[[:digit:]]", sep = ""),
names(glbObsAll), value = TRUE))
indepVars <- myadjust_interaction_feats(indepVars)
ret_lst <- myfit_mdl(mdl_specs_lst = myinit_mdl_specs_lst(mdl_specs_lst = list(
id.prefix = "Max.cor.Y.Time.Lag",
type = glb_model_type,
tune.df = glbMdlTuneParams,
trainControl.method = "repeatedcv",
trainControl.number = glb_rcv_n_folds, trainControl.repeats = glb_rcv_n_repeats,
trainControl.classProbs = glb_is_classification,
trainControl.summaryFunction = glbMdlMetricSummaryFn,
train.metric = glbMdlMetricSummary,
train.maximize = glbMdlMetricMaximize,
train.method = "glmnet")),
indep_vars = indepVars,
rsp_var = glb_rsp_var,
fit_df = glbObsFit, OOB_df = glbObsOOB)
}
if (length(glbFeatsText) > 0) {
fit.models_0_chunk_df <- myadd_chunk(fit.models_0_chunk_df,
paste0("fit.models_0_", "Txt.*"), major.inc = FALSE,
label.minor = "glmnet")
indepVars <- c(max_cor_y_x_vars)
for (txtFeat in names(glbFeatsText))
indepVars <- union(indepVars,
grep(paste(str_to_upper(substr(txtFeat, 1, 1)), "\\.(?!([T|P]\\.))", sep = ""),
names(glbObsAll), perl = TRUE, value = TRUE))
indepVars <- myadjust_interaction_feats(indepVars)
ret_lst <- myfit_mdl(mdl_specs_lst = myinit_mdl_specs_lst(mdl_specs_lst = list(
id.prefix = "Max.cor.Y.Text.nonTP",
type = glb_model_type,
tune.df = glbMdlTuneParams,
trainControl.method = "repeatedcv",
trainControl.number = glb_rcv_n_folds, trainControl.repeats = glb_rcv_n_repeats,
trainControl.classProbs = glb_is_classification,
trainControl.summaryFunction = glbMdlMetricSummaryFn,
trainControl.allowParallel = glbMdlAllowParallel,
train.metric = glbMdlMetricSummary,
train.maximize = glbMdlMetricMaximize,
train.method = "glmnet")),
indep_vars = indepVars,
rsp_var = glb_rsp_var,
fit_df = glbObsFit, OOB_df = glbObsOOB)
indepVars <- c(max_cor_y_x_vars)
for (txtFeat in names(glbFeatsText))
indepVars <- union(indepVars,
grep(paste(str_to_upper(substr(txtFeat, 1, 1)), "\\.T\\.", sep = ""),
names(glbObsAll), perl = TRUE, value = TRUE))
indepVars <- myadjust_interaction_feats(indepVars)
ret_lst <- myfit_mdl(mdl_specs_lst = myinit_mdl_specs_lst(mdl_specs_lst = list(
id.prefix = "Max.cor.Y.Text.onlyT",
type = glb_model_type,
tune.df = glbMdlTuneParams,
trainControl.method = "repeatedcv",
trainControl.number = glb_rcv_n_folds, trainControl.repeats = glb_rcv_n_repeats,
trainControl.classProbs = glb_is_classification,
trainControl.summaryFunction = glbMdlMetricSummaryFn,
train.metric = glbMdlMetricSummary,
train.maximize = glbMdlMetricMaximize,
train.method = "glmnet")),
indep_vars = indepVars,
rsp_var = glb_rsp_var,
fit_df = glbObsFit, OOB_df = glbObsOOB)
indepVars <- c(max_cor_y_x_vars)
for (txtFeat in names(glbFeatsText))
indepVars <- union(indepVars,
grep(paste(str_to_upper(substr(txtFeat, 1, 1)), "\\.P\\.", sep = ""),
names(glbObsAll), perl = TRUE, value = TRUE))
indepVars <- myadjust_interaction_feats(indepVars)
ret_lst <- myfit_mdl(mdl_specs_lst = myinit_mdl_specs_lst(mdl_specs_lst = list(
id.prefix = "Max.cor.Y.Text.onlyP",
type = glb_model_type,
tune.df = glbMdlTuneParams,
trainControl.method = "repeatedcv",
trainControl.number = glb_rcv_n_folds, trainControl.repeats = glb_rcv_n_repeats,
trainControl.classProbs = glb_is_classification,
trainControl.summaryFunction = glbMdlMetricSummaryFn,
trainControl.allowParallel = glbMdlAllowParallel,
train.metric = glbMdlMetricSummary,
train.maximize = glbMdlMetricMaximize,
train.method = "glmnet")),
indep_vars = indepVars,
rsp_var = glb_rsp_var,
fit_df = glbObsFit, OOB_df = glbObsOOB)
}
# Interactions.High.cor.Y
if (length(int_feats <- setdiff(setdiff(unique(glb_feats_df$cor.high.X), NA),
subset(glb_feats_df, nzv)$id)) > 0) {
fit.models_0_chunk_df <- myadd_chunk(fit.models_0_chunk_df,
paste0("fit.models_0_", "Interact.High.cor.Y"), major.inc = FALSE,
label.minor = "glmnet")
ret_lst <- myfit_mdl(mdl_specs_lst=myinit_mdl_specs_lst(mdl_specs_lst=list(
id.prefix="Interact.High.cor.Y",
type=glb_model_type, trainControl.method="repeatedcv",
trainControl.number=glb_rcv_n_folds, trainControl.repeats=glb_rcv_n_repeats,
trainControl.classProbs = glb_is_classification,
trainControl.summaryFunction = glbMdlMetricSummaryFn,
trainControl.allowParallel = glbMdlAllowParallel,
train.metric = glbMdlMetricSummary,
train.maximize = glbMdlMetricMaximize,
train.method="glmnet")),
indep_vars=c(max_cor_y_x_vars, paste(max_cor_y_x_vars[1], int_feats, sep=":")),
rsp_var=glb_rsp_var,
fit_df=glbObsFit, OOB_df=glbObsOOB)
} ## label step_major step_minor label_minor
## 4 fit.models_0_Max.cor.Y.rcv.*X* 1 3 glmnet
## 5 fit.models_0_Interact.High.cor.Y 1 4 glmnet
## bgn end elapsed
## 4 140.137 149.278 9.142
## 5 149.279 NA NA
## [1] "myfit_mdl: enter: 0.000000 secs"
## [1] "fitting model: Interact.High.cor.Y##rcv#glmnet"
## [1] " indep_vars: nImgs.cut.fctr,resY.min,nImgs.cut.fctr:nImgs.cut.fctr,nImgs.cut.fctr:resX.mad.log1p,nImgs.cut.fctr:nImgs.log1p,nImgs.cut.fctr:resY.mean.log1p,nImgs.cut.fctr:resY.mean.root2,nImgs.cut.fctr:resXY.mad,nImgs.cut.fctr:resXY.mad.nexp,nImgs.cut.fctr:resX.mean,nImgs.cut.fctr:resX.mean.nexp,nImgs.cut.fctr:resX.min,nImgs.cut.fctr:resXY.min,nImgs.cut.fctr:resY.min"
## [1] "myfit_mdl: setup complete: 0.683000 secs"
## Aggregating results
## Selecting tuning parameters
## Fitting alpha = 0.775, lambda = 4.65e-05 on full training set
## [1] "myfit_mdl: train complete: 23.119000 secs"## Warning in myfit_mdl(mdl_specs_lst = myinit_mdl_specs_lst(mdl_specs_lst
## = list(id.prefix = "Interact.High.cor.Y", : model's bestTune found at an
## extreme of tuneGrid for parameter: lambda
## Length Class Mode
## a0 69 -none- numeric
## beta 3243 dgCMatrix S4
## df 69 -none- numeric
## dim 2 -none- numeric
## lambda 69 -none- numeric
## dev.ratio 69 -none- numeric
## nulldev 1 -none- numeric
## npasses 1 -none- numeric
## jerr 1 -none- numeric
## offset 1 -none- logical
## classnames 2 -none- character
## call 5 -none- call
## nobs 1 -none- numeric
## lambdaOpt 1 -none- numeric
## xNames 47 -none- character
## problemType 1 -none- character
## tuneValue 2 data.frame list
## obsLevels 2 -none- character
## [1] "min lambda > lambdaOpt:"
## (Intercept)
## 6.115004e+00
## nImgs.cut.fctr(32,60]
## -1.088896e+01
## nImgs.cut.fctr(120,3e+03]
## 9.149823e-01
## resY.min
## -1.058372e-04
## nImgs.cut.fctr(0,32]:nImgs.log1p
## 7.957820e-01
## nImgs.cut.fctr(32,60]:nImgs.log1p
## 1.712224e-01
## nImgs.cut.fctr(60,120]:nImgs.log1p
## -4.345255e-01
## nImgs.cut.fctr(120,3e+03]:nImgs.log1p
## -1.625661e-01
## nImgs.cut.fctr(0,32]:resX.mad.log1p
## 1.293689e-02
## nImgs.cut.fctr(32,60]:resX.mad.log1p
## 3.942754e-02
## nImgs.cut.fctr(120,3e+03]:resX.mad.log1p
## 1.448236e-01
## nImgs.cut.fctr(0,32]:resX.mean
## -7.688121e-03
## nImgs.cut.fctr(32,60]:resX.mean
## 9.164700e-03
## nImgs.cut.fctr(60,120]:resX.mean
## -2.526216e-03
## nImgs.cut.fctr(120,3e+03]:resX.mean
## -2.297982e-02
## nImgs.cut.fctr(0,32]:resX.mean.nexp
## 9.900000e+35
## nImgs.cut.fctr(32,60]:resX.mean.nexp
## 9.900000e+35
## nImgs.cut.fctr(0,32]:resX.min
## 3.544555e-03
## nImgs.cut.fctr(32,60]:resX.min
## 3.062510e-03
## nImgs.cut.fctr(60,120]:resX.min
## -4.431001e-03
## nImgs.cut.fctr(120,3e+03]:resX.min
## 1.283089e-02
## nImgs.cut.fctr(0,32]:resXY.mad
## 1.366340e-05
## nImgs.cut.fctr(32,60]:resXY.mad
## -2.024373e-06
## nImgs.cut.fctr(60,120]:resXY.mad
## 1.076650e-05
## nImgs.cut.fctr(120,3e+03]:resXY.mad
## 2.648629e-05
## nImgs.cut.fctr(0,32]:resXY.mad.nexp
## 7.092943e-02
## nImgs.cut.fctr(32,60]:resXY.mad.nexp
## 3.051364e-01
## nImgs.cut.fctr(60,120]:resXY.mad.nexp
## 2.404053e-01
## nImgs.cut.fctr(120,3e+03]:resXY.mad.nexp
## 6.629458e-01
## nImgs.cut.fctr(0,32]:resXY.min
## -4.192213e-06
## nImgs.cut.fctr(32,60]:resXY.min
## -1.379372e-05
## nImgs.cut.fctr(60,120]:resXY.min
## 1.487104e-05
## nImgs.cut.fctr(120,3e+03]:resXY.min
## -1.299658e-05
## nImgs.cut.fctr(32,60]:resY.mean.log1p
## -2.320909e-01
## nImgs.cut.fctr(120,3e+03]:resY.mean.log1p
## 1.325613e-01
## nImgs.cut.fctr(0,32]:resY.mean.root2
## -2.987543e-01
## nImgs.cut.fctr(32,60]:resY.mean.root2
## 8.509761e-03
## nImgs.cut.fctr(60,120]:resY.mean.root2
## -1.253344e-01
## nImgs.cut.fctr(32,60]:resY.min
## 7.070070e-03
## nImgs.cut.fctr(60,120]:resY.min
## -4.026836e-03
## nImgs.cut.fctr(120,3e+03]:resY.min
## 2.738971e-03
## [1] "max lambda < lambdaOpt:"
## [1] "Feats mismatch between coefs_left & rght:"
## [1] "(Intercept)"
## [2] "nImgs.cut.fctr(32,60]"
## [3] "nImgs.cut.fctr(60,120]"
## [4] "nImgs.cut.fctr(120,3e+03]"
## [5] "resY.min"
## [6] "nImgs.cut.fctr(0,32]:nImgs.log1p"
## [7] "nImgs.cut.fctr(32,60]:nImgs.log1p"
## [8] "nImgs.cut.fctr(60,120]:nImgs.log1p"
## [9] "nImgs.cut.fctr(120,3e+03]:nImgs.log1p"
## [10] "nImgs.cut.fctr(0,32]:resX.mad.log1p"
## [11] "nImgs.cut.fctr(32,60]:resX.mad.log1p"
## [12] "nImgs.cut.fctr(60,120]:resX.mad.log1p"
## [13] "nImgs.cut.fctr(120,3e+03]:resX.mad.log1p"
## [14] "nImgs.cut.fctr(0,32]:resX.mean"
## [15] "nImgs.cut.fctr(32,60]:resX.mean"
## [16] "nImgs.cut.fctr(60,120]:resX.mean"
## [17] "nImgs.cut.fctr(120,3e+03]:resX.mean"
## [18] "nImgs.cut.fctr(0,32]:resX.mean.nexp"
## [19] "nImgs.cut.fctr(32,60]:resX.mean.nexp"
## [20] "nImgs.cut.fctr(60,120]:resX.mean.nexp"
## [21] "nImgs.cut.fctr(120,3e+03]:resX.mean.nexp"
## [22] "nImgs.cut.fctr(0,32]:resX.min"
## [23] "nImgs.cut.fctr(32,60]:resX.min"
## [24] "nImgs.cut.fctr(60,120]:resX.min"
## [25] "nImgs.cut.fctr(120,3e+03]:resX.min"
## [26] "nImgs.cut.fctr(0,32]:resXY.mad"
## [27] "nImgs.cut.fctr(32,60]:resXY.mad"
## [28] "nImgs.cut.fctr(60,120]:resXY.mad"
## [29] "nImgs.cut.fctr(120,3e+03]:resXY.mad"
## [30] "nImgs.cut.fctr(0,32]:resXY.mad.nexp"
## [31] "nImgs.cut.fctr(32,60]:resXY.mad.nexp"
## [32] "nImgs.cut.fctr(60,120]:resXY.mad.nexp"
## [33] "nImgs.cut.fctr(120,3e+03]:resXY.mad.nexp"
## [34] "nImgs.cut.fctr(0,32]:resXY.min"
## [35] "nImgs.cut.fctr(32,60]:resXY.min"
## [36] "nImgs.cut.fctr(60,120]:resXY.min"
## [37] "nImgs.cut.fctr(120,3e+03]:resXY.min"
## [38] "nImgs.cut.fctr(0,32]:resY.mean.log1p"
## [39] "nImgs.cut.fctr(32,60]:resY.mean.log1p"
## [40] "nImgs.cut.fctr(60,120]:resY.mean.log1p"
## [41] "nImgs.cut.fctr(120,3e+03]:resY.mean.log1p"
## [42] "nImgs.cut.fctr(0,32]:resY.mean.root2"
## [43] "nImgs.cut.fctr(32,60]:resY.mean.root2"
## [44] "nImgs.cut.fctr(60,120]:resY.mean.root2"
## [45] "nImgs.cut.fctr(120,3e+03]:resY.mean.root2"
## [46] "nImgs.cut.fctr(32,60]:resY.min"
## [47] "nImgs.cut.fctr(60,120]:resY.min"
## [48] "nImgs.cut.fctr(120,3e+03]:resY.min"
## [1] "myfit_mdl: train diagnostics complete: 23.726000 secs"
## Prediction
## Reference N Y
## N 25 475
## Y 8 494
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 5.179641e-01 3.412724e-02 4.865121e-01 5.493102e-01 5.009980e-01
## AccuracyPValue McnemarPValue
## 1.485851e-01 8.821190e-100
## Prediction
## Reference N Y
## N 7 490
## Y 2 499
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 5.070140e-01 1.013221e-02 4.755136e-01 5.384730e-01 5.020040e-01
## AccuracyPValue McnemarPValue
## 3.878866e-01 7.647039e-107
## [1] "myfit_mdl: predict complete: 26.607000 secs"
## id
## 1 Interact.High.cor.Y##rcv#glmnet
## feats
## 1 nImgs.cut.fctr,resY.min,nImgs.cut.fctr:nImgs.cut.fctr,nImgs.cut.fctr:resX.mad.log1p,nImgs.cut.fctr:nImgs.log1p,nImgs.cut.fctr:resY.mean.log1p,nImgs.cut.fctr:resY.mean.root2,nImgs.cut.fctr:resXY.mad,nImgs.cut.fctr:resXY.mad.nexp,nImgs.cut.fctr:resX.mean,nImgs.cut.fctr:resX.mean.nexp,nImgs.cut.fctr:resX.min,nImgs.cut.fctr:resXY.min,nImgs.cut.fctr:resY.min
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 25 22.427 1.069
## max.AUCpROC.fit max.Sens.fit max.Spec.fit max.AUCROCR.fit
## 1 0.6058048 0.614 0.5976096 0.642749
## opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.3 0.6716519 0.5448976
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit
## 1 0.4865121 0.5493102 0.08975773
## max.AUCpROC.OOB max.Sens.OOB max.Spec.OOB max.AUCROCR.OOB
## 1 0.5040041 0.5030181 0.50499 0.5140745
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.2 0.6697987 0.507014
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.4755136 0.538473 0.01013221
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.02621502 0.05242679
## [1] "myfit_mdl: exit: 26.621000 secs"# Low.cor.X
fit.models_0_chunk_df <- myadd_chunk(fit.models_0_chunk_df,
paste0("fit.models_0_", "Low.cor.X"), major.inc = FALSE,
label.minor = "glmnet")## label step_major step_minor label_minor
## 5 fit.models_0_Interact.High.cor.Y 1 4 glmnet
## 6 fit.models_0_Low.cor.X 1 5 glmnet
## bgn end elapsed
## 5 149.279 175.936 26.657
## 6 175.936 NA NAindep_vars <- subset(glb_feats_df, is.na(cor.high.X) & !nzv &
(exclude.as.feat != 1))[, "id"]
indep_vars <- myadjust_interaction_feats(indep_vars)
ret_lst <- myfit_mdl(mdl_specs_lst = myinit_mdl_specs_lst(mdl_specs_lst = list(
id.prefix = "Low.cor.X",
type = glb_model_type,
tune.df = glbMdlTuneParams,
trainControl.method = "repeatedcv",
trainControl.number = glb_rcv_n_folds, trainControl.repeats = glb_rcv_n_repeats,
trainControl.classProbs = glb_is_classification,
trainControl.summaryFunction = glbMdlMetricSummaryFn,
trainControl.allowParallel = glbMdlAllowParallel,
train.metric = glbMdlMetricSummary,
train.maximize = glbMdlMetricMaximize,
train.method = "glmnet")),
indep_vars = indep_vars, rsp_var = glb_rsp_var,
fit_df = glbObsFit, OOB_df = glbObsOOB)## [1] "myfit_mdl: enter: 0.000000 secs"
## [1] "fitting model: Low.cor.X##rcv#glmnet"
## [1] " indep_vars: nImgs.cut.fctr,.pos,resX.mad.log1p,resXY.mad.nexp,resY.mean.log1p,resY.mad.nexp,resY.mad,resXY.max.log1p,resXY.max.root2,resXY.max,resY.mad.root2,resY.mad.log1p,nImgs.nexp,resXY.mean.log1p,resXY.mean.root2,.rnorm,resXY.mean,resXY.mad,nImgs,resX.mean,resX.min.nexp,resX.mean.nexp,resY.min.nexp,resX.min,resY.min"
## [1] "myfit_mdl: setup complete: 0.674000 secs"
## Aggregating results
## Selecting tuning parameters
## Fitting alpha = 0.1, lambda = 2.27e-05 on full training set
## [1] "myfit_mdl: train complete: 7.163000 secs"## Warning in myfit_mdl(mdl_specs_lst = myinit_mdl_specs_lst(mdl_specs_lst
## = list(id.prefix = "Low.cor.X", : model's bestTune found at an extreme of
## tuneGrid for parameter: alpha## Warning in myfit_mdl(mdl_specs_lst = myinit_mdl_specs_lst(mdl_specs_lst
## = list(id.prefix = "Low.cor.X", : model's bestTune found at an extreme of
## tuneGrid for parameter: lambda
## Length Class Mode
## a0 100 -none- numeric
## beta 2700 dgCMatrix S4
## df 100 -none- numeric
## dim 2 -none- numeric
## lambda 100 -none- numeric
## dev.ratio 100 -none- numeric
## nulldev 1 -none- numeric
## npasses 1 -none- numeric
## jerr 1 -none- numeric
## offset 1 -none- logical
## classnames 2 -none- character
## call 5 -none- call
## nobs 1 -none- numeric
## lambdaOpt 1 -none- numeric
## xNames 27 -none- character
## problemType 1 -none- character
## tuneValue 2 data.frame list
## obsLevels 2 -none- character
## [1] "min lambda > lambdaOpt:"
## (Intercept) .pos
## -5.137400e+02 9.469314e-05
## .rnorm nImgs
## -6.696257e-02 -6.025466e-04
## nImgs.cut.fctr(32,60] nImgs.cut.fctr(60,120]
## 3.055406e-01 3.605103e-01
## nImgs.cut.fctr(120,3e+03] nImgs.nexp
## 7.032784e-01 3.883642e+01
## resX.mad.log1p resX.mean
## 2.946340e-02 1.789300e-02
## resX.mean.nexp resX.min
## 9.900000e+35 1.896118e-03
## resX.min.nexp resXY.mad
## -2.074840e+27 -6.363773e-08
## resXY.mad.nexp resXY.max
## 2.331664e-01 -1.796879e-04
## resXY.max.log1p resXY.max.root2
## 3.256226e+01 3.115691e-02
## resXY.mean resXY.mean.log1p
## -1.086317e-04 8.228750e+00
## resXY.mean.root2 resY.mad
## -4.986359e-03 1.063081e-02
## resY.mad.log1p resY.mad.nexp
## 1.163426e+00 1.078418e+00
## resY.mad.root2 resY.mean.log1p
## -5.324178e-01 8.403533e+00
## resY.min resY.min.nexp
## -8.449056e-04 -7.675900e+21
## [1] "max lambda < lambdaOpt:"
## [1] "Feats mismatch between coefs_left & rght:"
## [1] "(Intercept)" ".pos"
## [3] ".rnorm" "nImgs"
## [5] "nImgs.cut.fctr(32,60]" "nImgs.cut.fctr(60,120]"
## [7] "nImgs.cut.fctr(120,3e+03]" "nImgs.nexp"
## [9] "resX.mad.log1p" "resX.mean"
## [11] "resX.mean.nexp" "resX.min"
## [13] "resX.min.nexp" "resXY.mad"
## [15] "resXY.mad.nexp" "resXY.max"
## [17] "resXY.max.log1p" "resXY.max.root2"
## [19] "resXY.mean" "resXY.mean.log1p"
## [21] "resXY.mean.root2" "resY.mad"
## [23] "resY.mad.log1p" "resY.mad.nexp"
## [25] "resY.mad.root2" "resY.mean.log1p"
## [27] "resY.min" "resY.min.nexp"
## [1] "myfit_mdl: train diagnostics complete: 7.771000 secs"
## Prediction
## Reference N Y
## N 13 487
## Y 1 501
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 5.129741e-01 2.405454e-02 4.815272e-01 5.443444e-01 5.009980e-01
## AccuracyPValue McnemarPValue
## 2.337476e-01 7.772075e-107
## Prediction
## Reference N Y
## N 0 497
## Y 0 501
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 5.020040e-01 0.000000e+00 4.705156e-01 5.334806e-01 5.020040e-01
## AccuracyPValue McnemarPValue
## 5.126421e-01 1.162632e-109
## [1] "myfit_mdl: predict complete: 10.810000 secs"
## id
## 1 Low.cor.X##rcv#glmnet
## feats
## 1 nImgs.cut.fctr,.pos,resX.mad.log1p,resXY.mad.nexp,resY.mean.log1p,resY.mad.nexp,resY.mad,resXY.max.log1p,resXY.max.root2,resXY.max,resY.mad.root2,resY.mad.log1p,nImgs.nexp,resXY.mean.log1p,resXY.mean.root2,.rnorm,resXY.mean,resXY.mad,nImgs,resX.mean,resX.min.nexp,resX.mean.nexp,resY.min.nexp,resX.min,resY.min
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 25 6.475 0.633
## max.AUCpROC.fit max.Sens.fit max.Spec.fit max.AUCROCR.fit
## 1 0.5638247 0.54 0.5876494 0.5929761
## opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.3 0.6724832 0.5265996
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit
## 1 0.4815272 0.5443444 0.05311189
## max.AUCpROC.OOB max.Sens.OOB max.Spec.OOB max.AUCROCR.OOB
## 1 0.4978875 0.4708249 0.5249501 0.5012952
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0 0.6684456 0.502004
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.4705156 0.5334806 0
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.0195775 0.03917575
## [1] "myfit_mdl: exit: 10.825000 secs"fit.models_0_chunk_df <-
myadd_chunk(fit.models_0_chunk_df, "fit.models_0_end", major.inc = FALSE,
label.minor = "teardown")## label step_major step_minor label_minor bgn end
## 6 fit.models_0_Low.cor.X 1 5 glmnet 175.936 186.81
## 7 fit.models_0_end 1 6 teardown 186.811 NA
## elapsed
## 6 10.874
## 7 NArm(ret_lst)
glb_chunks_df <- myadd_chunk(glb_chunks_df, "fit.models", major.inc = FALSE)## label step_major step_minor label_minor bgn end elapsed
## 16 fit.models 8 0 0 132.969 186.823 53.854
## 17 fit.models 8 1 1 186.824 NA NAfit.models_1_chunk_df <- myadd_chunk(NULL, "fit.models_1_bgn", label.minor = "setup")## label step_major step_minor label_minor bgn end elapsed
## 1 fit.models_1_bgn 1 0 setup 190.945 NA NA# refactor code for outliers / ensure all model runs exclude outliers in this chunk ???
#stop(here"); glb2Sav(); all.equal(glb_models_df, sav_models_df)
topindep_var <- NULL; interact_vars <- NULL;
for (mdl_id_pfx in names(glbMdlFamilies)) {
fit.models_1_chunk_df <-
myadd_chunk(fit.models_1_chunk_df, paste0("fit.models_1_", mdl_id_pfx),
major.inc = FALSE, label.minor = "setup")
indep_vars <- NULL;
if (grepl("\\.Interact", mdl_id_pfx)) {
if (is.null(topindep_var) && is.null(interact_vars)) {
# select best glmnet model upto now
dsp_models_df <- orderBy(model_sel_frmla <- get_model_sel_frmla(),
glb_models_df)
dsp_models_df <- subset(dsp_models_df,
grepl(".glmnet", id, fixed = TRUE))
bst_mdl_id <- dsp_models_df$id[1]
mdl_id_pfx <-
paste(c(head(unlist(strsplit(bst_mdl_id, "[.]")), -1), "Interact"),
collapse=".")
# select important features
if (is.null(bst_featsimp_df <-
myget_feats_importance(glb_models_lst[[bst_mdl_id]]))) {
warning("Base model for RFE.Interact: ", bst_mdl_id,
" has no important features")
next
}
topindep_ix <- 1
while (is.null(topindep_var) && (topindep_ix <= nrow(bst_featsimp_df))) {
topindep_var <- row.names(bst_featsimp_df)[topindep_ix]
if (grepl(".fctr", topindep_var, fixed=TRUE))
topindep_var <-
paste0(unlist(strsplit(topindep_var, ".fctr"))[1], ".fctr")
if (topindep_var %in% names(glbFeatsInteractionOnly)) {
topindep_var <- NULL; topindep_ix <- topindep_ix + 1
} else break
}
# select features with importance > max(10, importance of .rnorm) & is not highest
# combine factor dummy features to just the factor feature
if (length(pos_rnorm <-
grep(".rnorm", row.names(bst_featsimp_df), fixed=TRUE)) > 0)
imp_rnorm <- bst_featsimp_df[pos_rnorm, 1] else
imp_rnorm <- NA
imp_cutoff <- max(10, imp_rnorm, na.rm=TRUE)
interact_vars <-
tail(row.names(subset(bst_featsimp_df,
imp > imp_cutoff)), -1)
if (length(interact_vars) > 0) {
interact_vars <-
myadjust_interaction_feats(myextract_actual_feats(interact_vars))
interact_vars <-
interact_vars[!grepl(topindep_var, interact_vars, fixed=TRUE)]
}
### bid0_sp only
# interact_vars <- c(
# "biddable", "D.ratio.sum.TfIdf.wrds.n", "D.TfIdf.sum.stem.stop.Ratio", "D.sum.TfIdf",
# "D.TfIdf.sum.post.stop", "D.TfIdf.sum.post.stem", "D.ratio.wrds.stop.n.wrds.n", "D.chrs.uppr.n.log",
# "D.chrs.n.log", "color.fctr"
# # , "condition.fctr", "prdl.my.descr.fctr"
# )
# interact_vars <- setdiff(interact_vars, c("startprice.dgt2.is9", "color.fctr"))
###
indep_vars <- myextract_actual_feats(row.names(bst_featsimp_df))
indep_vars <- setdiff(indep_vars, topindep_var)
if (length(interact_vars) > 0) {
indep_vars <-
setdiff(indep_vars, myextract_actual_feats(interact_vars))
indep_vars <- c(indep_vars,
paste(topindep_var, setdiff(interact_vars, topindep_var),
sep = "*"))
} else indep_vars <- union(indep_vars, topindep_var)
}
}
if (is.null(indep_vars))
indep_vars <- glb_mdl_feats_lst[[mdl_id_pfx]]
if (is.null(indep_vars) && grepl("RFE\\.", mdl_id_pfx))
indep_vars <- myextract_actual_feats(predictors(rfe_fit_results))
if (is.null(indep_vars))
indep_vars <- subset(glb_feats_df, !nzv & (exclude.as.feat != 1))[, "id"]
if ((length(indep_vars) == 1) && (grepl("^%<d-%", indep_vars))) {
indep_vars <-
eval(parse(text = str_trim(unlist(strsplit(indep_vars, "%<d-%"))[2])))
}
indep_vars <- myadjust_interaction_feats(indep_vars)
if (grepl("\\.Interact", mdl_id_pfx)) {
# if (method != tail(unlist(strsplit(bst_mdl_id, "[.]")), 1)) next
if (is.null(glbMdlFamilies[[mdl_id_pfx]])) {
if (!is.null(glbMdlFamilies[["Best.Interact"]]))
glbMdlFamilies[[mdl_id_pfx]] <-
glbMdlFamilies[["Best.Interact"]]
}
}
if (!is.null(glbObsFitOutliers[[mdl_id_pfx]])) {
fitobs_df <- glbObsFit[!(glbObsFit[, glbFeatsId] %in%
glbObsFitOutliers[[mdl_id_pfx]]), ]
print(sprintf("Outliers removed: %d", nrow(glbObsFit) - nrow(fitobs_df)))
print(setdiff(glbObsFit[, glbFeatsId], fitobs_df[, glbFeatsId]))
} else fitobs_df <- glbObsFit
if (is.null(glbMdlFamilies[[mdl_id_pfx]]))
mdl_methods <- glbMdlMethods else
mdl_methods <- glbMdlFamilies[[mdl_id_pfx]]
for (method in mdl_methods) {
if (method %in% c("rpart", "rf")) {
# rpart: fubar's the tree
# rf: skip the scenario w/ .rnorm for speed
indep_vars <- setdiff(indep_vars, c(".rnorm"))
#mdl_id <- paste0(mdl_id_pfx, ".no.rnorm")
}
fit.models_1_chunk_df <- myadd_chunk(fit.models_1_chunk_df,
paste0("fit.models_1_", mdl_id_pfx), major.inc = FALSE,
label.minor = method)
ret_lst <-
myfit_mdl(mdl_specs_lst = myinit_mdl_specs_lst(mdl_specs_lst = list(
id.prefix = mdl_id_pfx,
type = glb_model_type,
tune.df = glbMdlTuneParams,
trainControl.method = "repeatedcv", # or "none" if nominalWorkflow is crashing
trainControl.number = glb_rcv_n_folds,
trainControl.repeats = glb_rcv_n_repeats,
trainControl.classProbs = glb_is_classification,
trainControl.summaryFunction = glbMdlMetricSummaryFn,
trainControl.allowParallel = glbMdlAllowParallel,
train.metric = glbMdlMetricSummary,
train.maximize = glbMdlMetricMaximize,
train.method = method)),
indep_vars = indep_vars, rsp_var = glb_rsp_var,
fit_df = fitobs_df, OOB_df = glbObsOOB)
# ntv_mdl <- glmnet(x = as.matrix(
# fitobs_df[, indep_vars]),
# y = as.factor(as.character(
# fitobs_df[, glb_rsp_var])),
# family = "multinomial")
# bgn = 1; end = 100;
# ntv_mdl <- glmnet(x = as.matrix(
# subset(fitobs_df, pop.fctr != "crypto")[bgn:end, indep_vars]),
# y = as.factor(as.character(
# subset(fitobs_df, pop.fctr != "crypto")[bgn:end, glb_rsp_var])),
# family = "multinomial")
}
}## label step_major step_minor label_minor bgn end
## 1 fit.models_1_bgn 1 0 setup 190.945 190.955
## 2 fit.models_1_All.X 1 1 setup 190.956 NA
## elapsed
## 1 0.01
## 2 NA
## label step_major step_minor label_minor bgn end
## 2 fit.models_1_All.X 1 1 setup 190.956 190.963
## 3 fit.models_1_All.X 1 2 glmnet 190.964 NA
## elapsed
## 2 0.007
## 3 NA
## [1] "myfit_mdl: enter: 0.000000 secs"
## [1] "fitting model: All.X##rcv#glmnet"
## [1] " indep_vars: nImgs.cut.fctr,nImgs.log1p,.pos,resX.mad.log1p,resX.mad.root2,resX.mad,resXY.mad.nexp,nImgs.root2,resY.mean.log1p,resY.mean.root2,resY.mean,resY.mad.nexp,resY.mad,resXY.max.log1p,resXY.max.root2,resXY.max,resY.mad.root2,resY.mad.log1p,nImgs.nexp,resXY.mean.log1p,resXY.mean.root2,.rnorm,resXY.mean,resXY.mad.root2,resXY.mad,resX.mad.nexp,resXY.mad.log1p,nImgs,resX.mean.log1p,resX.mean.root2,resX.mean,resX.min.nexp,resX.mean.nexp,resY.mean.nexp,resY.min.nexp,resX.min.log1p,resX.min.root2,resX.min,resXY.min.log1p,resXY.min.root2,resY.min.log1p,resY.min.root2,resXY.min,resY.min"
## [1] "myfit_mdl: setup complete: 0.710000 secs"
## Aggregating results
## Selecting tuning parameters
## Fitting alpha = 0.1, lambda = 2.45e-05 on full training set
## [1] "myfit_mdl: train complete: 11.771000 secs"## Warning in myfit_mdl(mdl_specs_lst = myinit_mdl_specs_lst(mdl_specs_lst
## = list(id.prefix = mdl_id_pfx, : model's bestTune found at an extreme of
## tuneGrid for parameter: alpha## Warning in myfit_mdl(mdl_specs_lst = myinit_mdl_specs_lst(mdl_specs_lst
## = list(id.prefix = mdl_id_pfx, : model's bestTune found at an extreme of
## tuneGrid for parameter: lambda
## Length Class Mode
## a0 100 -none- numeric
## beta 4600 dgCMatrix S4
## df 100 -none- numeric
## dim 2 -none- numeric
## lambda 100 -none- numeric
## dev.ratio 100 -none- numeric
## nulldev 1 -none- numeric
## npasses 1 -none- numeric
## jerr 1 -none- numeric
## offset 1 -none- logical
## classnames 2 -none- character
## call 5 -none- call
## nobs 1 -none- numeric
## lambdaOpt 1 -none- numeric
## xNames 46 -none- character
## problemType 1 -none- character
## tuneValue 2 data.frame list
## obsLevels 2 -none- character
## [1] "min lambda > lambdaOpt:"
## (Intercept) .pos
## -5.322001e+02 1.037459e-04
## .rnorm nImgs
## -6.769060e-02 1.521544e-04
## nImgs.cut.fctr(32,60] nImgs.cut.fctr(60,120]
## 1.309604e-01 6.523010e-02
## nImgs.cut.fctr(120,3e+03] nImgs.log1p
## 3.460511e-01 5.930509e-01
## nImgs.nexp nImgs.root2
## 5.294506e+01 -1.033055e-01
## resX.mad resX.mad.log1p
## -7.462293e-03 3.347949e-01
## resX.mad.nexp resX.mad.root2
## 8.597647e-01 1.847927e-02
## resX.mean resX.mean.log1p
## -2.908604e-02 1.718262e+01
## resX.mean.nexp resX.mean.root2
## 9.900000e+35 -1.044154e-01
## resX.min resX.min.log1p
## 1.111873e-02 -9.111545e+00
## resX.min.nexp resX.min.root2
## -3.246667e+27 7.845843e-01
## resXY.mad resXY.mad.log1p
## 2.701393e-06 1.155758e-01
## resXY.mad.nexp resXY.mad.root2
## 1.030157e+00 -2.916969e-03
## resXY.max resXY.max.log1p
## -1.265518e-04 3.573686e+01
## resXY.max.root2 resXY.mean
## -3.382568e-02 -1.511560e-04
## resXY.mean.log1p resXY.mean.root2
## 1.984867e+01 -6.256731e-03
## resXY.min resXY.min.log1p
## -3.496593e-05 1.785193e+00
## resXY.min.root2 resY.mad
## 8.746218e-03 1.065394e-02
## resY.mad.log1p resY.mad.nexp
## 1.160382e+00 1.082350e+00
## resY.mad.root2 resY.mean
## -5.316544e-01 7.943836e-02
## resY.mean.log1p resY.mean.nexp
## -2.938950e+01 9.900000e+35
## resY.mean.root2 resY.min
## 1.030505e-01 9.792523e-04
## resY.min.log1p resY.min.nexp
## -2.872098e+00 -6.257821e+21
## resY.min.root2
## 3.282024e-01
## [1] "max lambda < lambdaOpt:"
## [1] "Feats mismatch between coefs_left & rght:"
## [1] "(Intercept)" ".pos"
## [3] ".rnorm" "nImgs"
## [5] "nImgs.cut.fctr(32,60]" "nImgs.cut.fctr(60,120]"
## [7] "nImgs.cut.fctr(120,3e+03]" "nImgs.log1p"
## [9] "nImgs.nexp" "nImgs.root2"
## [11] "resX.mad" "resX.mad.log1p"
## [13] "resX.mad.nexp" "resX.mad.root2"
## [15] "resX.mean" "resX.mean.log1p"
## [17] "resX.mean.nexp" "resX.mean.root2"
## [19] "resX.min" "resX.min.log1p"
## [21] "resX.min.nexp" "resX.min.root2"
## [23] "resXY.mad" "resXY.mad.log1p"
## [25] "resXY.mad.nexp" "resXY.mad.root2"
## [27] "resXY.max" "resXY.max.log1p"
## [29] "resXY.max.root2" "resXY.mean"
## [31] "resXY.mean.log1p" "resXY.mean.root2"
## [33] "resXY.min" "resXY.min.log1p"
## [35] "resXY.min.root2" "resY.mad"
## [37] "resY.mad.log1p" "resY.mad.nexp"
## [39] "resY.mad.root2" "resY.mean"
## [41] "resY.mean.log1p" "resY.mean.nexp"
## [43] "resY.mean.root2" "resY.min"
## [45] "resY.min.log1p" "resY.min.nexp"
## [47] "resY.min.root2"
## [1] "myfit_mdl: train diagnostics complete: 12.473000 secs"
## Prediction
## Reference N Y
## N 32 468
## Y 4 498
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 5.289421e-01 5.613565e-02 4.974896e-01 5.602241e-01 5.009980e-01
## AccuracyPValue McnemarPValue
## 4.111828e-02 8.917990e-101
## Prediction
## Reference N Y
## N 0 497
## Y 0 501
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 5.020040e-01 0.000000e+00 4.705156e-01 5.334806e-01 5.020040e-01
## AccuracyPValue McnemarPValue
## 5.126421e-01 1.162632e-109
## [1] "myfit_mdl: predict complete: 16.090000 secs"
## id
## 1 All.X##rcv#glmnet
## feats
## 1 nImgs.cut.fctr,nImgs.log1p,.pos,resX.mad.log1p,resX.mad.root2,resX.mad,resXY.mad.nexp,nImgs.root2,resY.mean.log1p,resY.mean.root2,resY.mean,resY.mad.nexp,resY.mad,resXY.max.log1p,resXY.max.root2,resXY.max,resY.mad.root2,resY.mad.log1p,nImgs.nexp,resXY.mean.log1p,resXY.mean.root2,.rnorm,resXY.mean,resXY.mad.root2,resXY.mad,resX.mad.nexp,resXY.mad.log1p,nImgs,resX.mean.log1p,resX.mean.root2,resX.mean,resX.min.nexp,resX.mean.nexp,resY.mean.nexp,resY.min.nexp,resX.min.log1p,resX.min.root2,resX.min,resXY.min.log1p,resXY.min.root2,resY.min.log1p,resY.min.root2,resXY.min,resY.min
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 25 11.043 1.193
## max.AUCpROC.fit max.Sens.fit max.Spec.fit max.AUCROCR.fit
## 1 0.5678247 0.548 0.5876494 0.6177371
## opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.3 0.6784741 0.5299353
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit
## 1 0.4974896 0.5602241 0.05971262
## max.AUCpROC.OOB max.Sens.OOB max.Spec.OOB max.AUCROCR.OOB
## 1 0.5108937 0.4788732 0.5429142 0.5204922
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0 0.6684456 0.502004
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.4705156 0.5334806 0
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.02333108 0.04665017
## [1] "myfit_mdl: exit: 16.105000 secs"
## label step_major step_minor label_minor bgn end
## 3 fit.models_1_All.X 1 2 glmnet 190.964 207.079
## 4 fit.models_1_All.X 1 3 glm 207.080 NA
## elapsed
## 3 16.115
## 4 NA
## [1] "myfit_mdl: enter: 0.000000 secs"
## [1] "fitting model: All.X##rcv#glm"
## [1] " indep_vars: nImgs.cut.fctr,nImgs.log1p,.pos,resX.mad.log1p,resX.mad.root2,resX.mad,resXY.mad.nexp,nImgs.root2,resY.mean.log1p,resY.mean.root2,resY.mean,resY.mad.nexp,resY.mad,resXY.max.log1p,resXY.max.root2,resXY.max,resY.mad.root2,resY.mad.log1p,nImgs.nexp,resXY.mean.log1p,resXY.mean.root2,.rnorm,resXY.mean,resXY.mad.root2,resXY.mad,resX.mad.nexp,resXY.mad.log1p,nImgs,resX.mean.log1p,resX.mean.root2,resX.mean,resX.min.nexp,resX.mean.nexp,resY.mean.nexp,resY.min.nexp,resX.min.log1p,resX.min.root2,resX.min,resXY.min.log1p,resXY.min.root2,resY.min.log1p,resY.min.root2,resXY.min,resY.min"
## [1] "myfit_mdl: setup complete: 0.689000 secs"
## Aggregating results
## Fitting final model on full training set
## [1] "myfit_mdl: train complete: 2.195000 secs"
## Warning in sqrt(crit * p * (1 - hh)/hh): NaNs produced## Warning in sqrt(crit * p * (1 - hh)/hh): NaNs produced
##
## Call:
## NULL
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.026 -1.143 0.266 1.125 1.788
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 2.708e+04 2.491e+04 1.087 0.2768
## .pos 1.096e-04 1.152e-04 0.951 0.3414
## .rnorm -7.722e-02 6.399e-02 -1.207 0.2275
## nImgs -6.547e-05 2.101e-03 -0.031 0.9751
## `nImgs.cut.fctr(32,60]` 1.116e-01 3.067e-01 0.364 0.7160
## `nImgs.cut.fctr(60,120]` 3.871e-02 4.372e-01 0.089 0.9294
## `nImgs.cut.fctr(120,3e+03]` 3.151e-01 6.030e-01 0.523 0.6012
## nImgs.log1p 5.186e-01 6.587e-01 0.787 0.4311
## nImgs.nexp -2.328e+00 3.413e+02 -0.007 0.9946
## nImgs.root2 -8.601e-02 1.606e-01 -0.535 0.5924
## resX.mad -2.205e-02 4.616e-02 -0.478 0.6329
## resX.mad.log1p -1.827e-01 1.841e+00 -0.099 0.9210
## resX.mad.nexp 8.272e-01 8.112e-01 1.020 0.3079
## resX.mad.root2 4.028e-01 1.259e+00 0.320 0.7491
## resX.mean 9.477e-01 8.155e+00 0.116 0.9075
## resX.mean.log1p 4.437e+02 3.494e+03 0.127 0.8989
## resX.mean.nexp 4.704e+153 Inf 0.000 1.0000
## resX.mean.root2 -8.156e+01 6.745e+02 -0.121 0.9038
## resX.min -1.323e-01 9.231e-02 -1.433 0.1518
## resX.min.log1p -4.177e+01 2.048e+01 -2.039 0.0414 *
## resX.min.nexp -4.471e+27 1.783e+28 -0.251 0.8020
## resX.min.root2 9.501e+00 5.524e+00 1.720 0.0855 .
## resXY.mad 4.489e-06 9.404e-05 0.048 0.9619
## resXY.mad.log1p 1.905e-01 1.115e+00 0.171 0.8644
## resXY.mad.nexp 1.496e+00 6.532e+00 0.229 0.8189
## resXY.mad.root2 -4.913e-03 4.524e-02 -0.109 0.9135
## resXY.max -4.066e-03 8.854e-03 -0.459 0.6461
## resXY.max.log1p -7.553e+02 1.902e+03 -0.397 0.6912
## resXY.max.root2 7.039e+00 1.643e+01 0.429 0.6683
## resXY.mean -1.351e-02 1.013e-02 -1.333 0.1825
## resXY.mean.log1p -2.276e+03 1.758e+03 -1.295 0.1954
## resXY.mean.root2 2.216e+01 1.689e+01 1.312 0.1894
## resXY.min 2.897e-05 5.299e-05 0.547 0.5846
## resXY.min.log1p 5.862e+00 3.358e+00 1.746 0.0808 .
## resXY.min.root2 -5.903e-02 5.392e-02 -1.095 0.2736
## resY.mad 5.300e-02 3.672e-02 1.443 0.1489
## resY.mad.log1p 2.907e+00 1.560e+00 1.863 0.0624 .
## resY.mad.nexp 1.627e+00 7.684e-01 2.118 0.0342 *
## resY.mad.root2 -1.715e+00 1.035e+00 -1.657 0.0975 .
## resY.mean -4.775e-01 3.396e+00 -0.141 0.8882
## resY.mean.log1p -2.850e+02 1.385e+03 -0.206 0.8370
## resY.mean.nexp -2.233e+129 3.157e+130 -0.071 0.9436
## resY.mean.root2 4.797e+01 2.739e+02 0.175 0.8610
## resY.min -6.743e-02 6.238e-02 -1.081 0.2798
## resY.min.log1p -1.617e+01 1.197e+01 -1.350 0.1769
## resY.min.nexp -8.983e+21 1.379e+22 -0.651 0.5147
## resY.min.root2 4.196e+00 3.476e+00 1.207 0.2274
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1389.1 on 1001 degrees of freedom
## Residual deviance: 1326.1 on 955 degrees of freedom
## AIC: 1420.1
##
## Number of Fisher Scoring iterations: 11
##
## [1] "myfit_mdl: train diagnostics complete: 3.024000 secs"
## Prediction
## Reference N Y
## N 43 457
## Y 10 492
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 5.339321e-01 6.619757e-02 5.024844e-01 5.651800e-01 5.009980e-01
## AccuracyPValue McnemarPValue
## 1.998543e-02 1.240552e-94
## Prediction
## Reference N Y
## N 0 497
## Y 0 501
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 5.020040e-01 0.000000e+00 4.705156e-01 5.334806e-01 5.020040e-01
## AccuracyPValue McnemarPValue
## 5.126421e-01 1.162632e-109
## [1] "myfit_mdl: predict complete: 6.602000 secs"
## id
## 1 All.X##rcv#glm
## feats
## 1 nImgs.cut.fctr,nImgs.log1p,.pos,resX.mad.log1p,resX.mad.root2,resX.mad,resXY.mad.nexp,nImgs.root2,resY.mean.log1p,resY.mean.root2,resY.mean,resY.mad.nexp,resY.mad,resXY.max.log1p,resXY.max.root2,resXY.max,resY.mad.root2,resY.mad.log1p,nImgs.nexp,resXY.mean.log1p,resXY.mean.root2,.rnorm,resXY.mean,resXY.mad.root2,resXY.mad,resX.mad.nexp,resXY.mad.log1p,nImgs,resX.mean.log1p,resX.mean.root2,resX.mean,resX.min.nexp,resX.mean.nexp,resY.mean.nexp,resY.min.nexp,resX.min.log1p,resX.min.root2,resX.min,resXY.min.log1p,resXY.min.root2,resY.min.log1p,resY.min.root2,resXY.min,resY.min
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 1 1.488 0.073
## max.AUCpROC.fit max.Sens.fit max.Spec.fit max.AUCROCR.fit
## 1 0.5797928 0.556 0.6035857 0.6320319
## opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.3 0.678153 0.5236096
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit
## 1 0.5024844 0.56518 0.04708344
## max.AUCpROC.OOB max.Sens.OOB max.Spec.OOB max.AUCROCR.OOB
## 1 0.5238678 0.4788732 0.5688623 0.5202713
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0 0.6684456 0.502004
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.4705156 0.5334806 0
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.01780093 0.03558361
## [1] "myfit_mdl: exit: 6.618000 secs"# Check if other preProcess methods improve model performance
fit.models_1_chunk_df <-
myadd_chunk(fit.models_1_chunk_df, "fit.models_1_preProc", major.inc = FALSE,
label.minor = "preProc")## label step_major step_minor label_minor bgn end
## 4 fit.models_1_All.X 1 3 glm 207.080 213.741
## 5 fit.models_1_preProc 1 4 preProc 213.742 NA
## elapsed
## 4 6.661
## 5 NAmdl_id <- orderBy(get_model_sel_frmla(), glb_models_df)[1, "id"]
indep_vars_vctr <- trim(unlist(strsplit(glb_models_df[glb_models_df$id == mdl_id,
"feats"], "[,]")))
method <- tail(unlist(strsplit(mdl_id, "[.]")), 1)
mdl_id_pfx <- paste0(head(unlist(strsplit(mdl_id, "[.]")), -1), collapse = ".")
if (!is.null(glbObsFitOutliers[[mdl_id_pfx]])) {
fitobs_df <- glbObsFit[!(glbObsFit[, glbFeatsId] %in%
glbObsFitOutliers[[mdl_id_pfx]]), ]
print(sprintf("Outliers removed: %d", nrow(glbObsFit) - nrow(fitobs_df)))
print(setdiff(glbObsFit[, glbFeatsId], fitobs_df[, glbFeatsId]))
} else fitobs_df <- glbObsFit
for (prePr in glb_preproc_methods) {
# The operations are applied in this order:
# Box-Cox/Yeo-Johnson transformation, centering, scaling, range, imputation, PCA, ICA then spatial sign.
ret_lst <- myfit_mdl(mdl_specs_lst=myinit_mdl_specs_lst(mdl_specs_lst=list(
id.prefix=mdl_id_pfx,
type=glb_model_type, tune.df=glbMdlTuneParams,
trainControl.method="repeatedcv",
trainControl.number=glb_rcv_n_folds,
trainControl.repeats=glb_rcv_n_repeats,
trainControl.classProbs = glb_is_classification,
trainControl.summaryFunction = glbMdlMetricSummaryFn,
train.metric = glbMdlMetricSummary,
train.maximize = glbMdlMetricMaximize,
train.method=method, train.preProcess=prePr)),
indep_vars=indep_vars_vctr, rsp_var=glb_rsp_var,
fit_df=fitobs_df, OOB_df=glbObsOOB)
}
# If (All|RFE).X.glm is less accurate than Low.Cor.X.glm
# check NA coefficients & filter appropriate terms in indep_vars_vctr
# if (method == "glm") {
# orig_glm <- glb_models_lst[[paste0(mdl_id, ".", model_method)]]$finalModel
# orig_glm <- glb_models_lst[["All.X.glm"]]$finalModel; print(summary(orig_glm))
# orig_glm <- glb_models_lst[["RFE.X.glm"]]$finalModel; print(summary(orig_glm))
# require(car)
# vif_orig_glm <- vif(orig_glm); print(vif_orig_glm)
# # if vif errors out with "there are aliased coefficients in the model"
# alias_orig_glm <- alias(orig_glm); alias_complete_orig_glm <- (alias_orig_glm$Complete > 0); alias_complete_orig_glm <- alias_complete_orig_glm[rowSums(alias_complete_orig_glm) > 0, colSums(alias_complete_orig_glm) > 0]; print(alias_complete_orig_glm)
# print(vif_orig_glm[!is.na(vif_orig_glm) & (vif_orig_glm == Inf)])
# print(which.max(vif_orig_glm))
# print(sort(vif_orig_glm[vif_orig_glm >= 1.0e+03], decreasing=TRUE))
# glbObsFit[c(1143, 3637, 3953, 4105), c("UniqueID", "Popular", "H.P.quandary", "Headline")]
# glb_feats_df[glb_feats_df$id %in% grep("[HSA]\\.chrs.n.log", glb_feats_df$id, value=TRUE) | glb_feats_df$cor.high.X %in% grep("[HSA]\\.chrs.n.log", glb_feats_df$id, value=TRUE), ]
# all.equal(glbObsAll$S.chrs.uppr.n.log, glbObsAll$A.chrs.uppr.n.log)
# cor(glbObsAll$S.T.herald, glbObsAll$S.T.tribun)
# mydspObs(Abstract.contains="[Dd]iar", cols=("Abstract"), all=TRUE)
# subset(glb_feats_df, cor.y.abs <= glb_feats_df[glb_feats_df$id == ".rnorm", "cor.y.abs"])
# corxx_mtrx <- cor(data.matrix(glbObsAll[, setdiff(names(glbObsAll), myfind_chr_cols_df(glbObsAll))]), use="pairwise.complete.obs"); abs_corxx_mtrx <- abs(corxx_mtrx); diag(abs_corxx_mtrx) <- 0
# which.max(abs_corxx_mtrx["S.T.tribun", ])
# abs_corxx_mtrx["A.npnct08.log", "S.npnct08.log"]
# step_glm <- step(orig_glm)
# }
# Since caret does not optimize rpart well
# if (method == "rpart")
# ret_lst <- myfit_mdl(mdl_id=paste0(mdl_id_pfx, ".cp.0"), model_method=method,
# indep_vars_vctr=indep_vars_vctr,
# model_type=glb_model_type,
# rsp_var=glb_rsp_var,
# fit_df=glbObsFit, OOB_df=glbObsOOB,
# n_cv_folds=0, tune_models_df=data.frame(parameter="cp", min=0.0, max=0.0, by=0.1))
# User specified
# Ensure at least 2 vars in each regression; else varImp crashes
# sav_models_lst <- glb_models_lst; sav_models_df <- glb_models_df; sav_featsimp_df <- glb_featsimp_df; all.equal(sav_featsimp_df, glb_featsimp_df)
# glb_models_lst <- sav_models_lst; glb_models_df <- sav_models_df; glm_featsimp_df <- sav_featsimp_df
# easier to exclude features
# require(gdata) # needed for trim
# mdl_id <- "";
# indep_vars_vctr <- head(subset(glb_models_df, grepl("All\\.X\\.", mdl_id), select=feats)
# , 1)[, "feats"]
# indep_vars_vctr <- trim(unlist(strsplit(indep_vars_vctr, "[,]")))
# indep_vars_vctr <- setdiff(indep_vars_vctr, ".rnorm")
# easier to include features
#stop(here"); sav_models_df <- glb_models_df; glb_models_df <- sav_models_df
# !_sp
# mdl_id <- "csm"; indep_vars_vctr <- c(NULL
# ,"prdline.my.fctr", "prdline.my.fctr:.clusterid.fctr"
# ,"prdline.my.fctr*biddable"
# #,"prdline.my.fctr*startprice.log"
# #,"prdline.my.fctr*startprice.diff"
# ,"prdline.my.fctr*condition.fctr"
# ,"prdline.my.fctr*D.terms.post.stop.n"
# #,"prdline.my.fctr*D.terms.post.stem.n"
# ,"prdline.my.fctr*cellular.fctr"
# # ,"<feat1>:<feat2>"
# )
# for (method in glbMdlMethods) {
# ret_lst <- myfit_mdl(mdl_id=mdl_id, model_method=method,
# indep_vars_vctr=indep_vars_vctr,
# model_type=glb_model_type,
# rsp_var=glb_rsp_var,
# fit_df=glbObsFit, OOB_df=glbObsOOB,
# n_cv_folds=glb_rcv_n_folds, tune_models_df=glbMdlTuneParams)
# csm_mdl_id <- paste0(mdl_id, ".", method)
# csm_featsimp_df <- myget_feats_importance(glb_models_lst[[paste0(mdl_id, ".",
# method)]]); print(head(csm_featsimp_df))
# }
###
# Ntv.1.lm <- lm(reformulate(indep_vars_vctr, glb_rsp_var), glbObsTrn); print(summary(Ntv.1.lm))
#glb_models_df[, "max.Accuracy.OOB", FALSE]
#varImp(glb_models_lst[["Low.cor.X.glm"]])
#orderBy(~ -Overall, varImp(glb_models_lst[["All.X.2.glm"]])$imp)
#orderBy(~ -Overall, varImp(glb_models_lst[["All.X.3.glm"]])$imp)
#glb_feats_df[grepl("npnct28", glb_feats_df$id), ]
# User specified bivariate models
# indep_vars_vctr_lst <- list()
# for (feat in setdiff(names(glbObsFit),
# union(glb_rsp_var, glbFeatsExclude)))
# indep_vars_vctr_lst[["feat"]] <- feat
# User specified combinatorial models
# indep_vars_vctr_lst <- list()
# combn_mtrx <- combn(c("<feat1_name>", "<feat2_name>", "<featn_name>"),
# <num_feats_to_choose>)
# for (combn_ix in 1:ncol(combn_mtrx))
# #print(combn_mtrx[, combn_ix])
# indep_vars_vctr_lst[[combn_ix]] <- combn_mtrx[, combn_ix]
# template for myfit_mdl
# rf is hard-coded in caret to recognize only Accuracy / Kappa evaluation metrics
# only for OOB in trainControl ?
# ret_lst <- myfit_mdl_fn(mdl_id=paste0(mdl_id_pfx, ""), model_method=method,
# indep_vars_vctr=indep_vars_vctr,
# rsp_var=glb_rsp_var,
# fit_df=glbObsFit, OOB_df=glbObsOOB,
# n_cv_folds=glb_rcv_n_folds, tune_models_df=glbMdlTuneParams,
# model_loss_mtrx=glbMdlMetric_terms,
# model_summaryFunction=glbMdlMetricSummaryFn,
# model_metric=glbMdlMetricSummary,
# model_metric_maximize=glbMdlMetricMaximize)
# Simplify a model
# fit_df <- glbObsFit; glb_mdl <- step(<complex>_mdl)
# Non-caret models
# rpart_area_mdl <- rpart(reformulate("Area", response=glb_rsp_var),
# data=glbObsFit, #method="class",
# control=rpart.control(cp=0.12),
# parms=list(loss=glbMdlMetric_terms))
# print("rpart_sel_wlm_mdl"); prp(rpart_sel_wlm_mdl)
#
print(glb_models_df)## id
## MFO###myMFO_classfr MFO###myMFO_classfr
## Random###myrandom_classfr Random###myrandom_classfr
## Max.cor.Y.rcv.1X1###glmnet Max.cor.Y.rcv.1X1###glmnet
## Max.cor.Y##rcv#rpart Max.cor.Y##rcv#rpart
## Interact.High.cor.Y##rcv#glmnet Interact.High.cor.Y##rcv#glmnet
## Low.cor.X##rcv#glmnet Low.cor.X##rcv#glmnet
## All.X##rcv#glmnet All.X##rcv#glmnet
## All.X##rcv#glm All.X##rcv#glm
## feats
## MFO###myMFO_classfr .rnorm
## Random###myrandom_classfr .rnorm
## Max.cor.Y.rcv.1X1###glmnet nImgs.cut.fctr,resY.min
## Max.cor.Y##rcv#rpart nImgs.cut.fctr,resY.min
## Interact.High.cor.Y##rcv#glmnet nImgs.cut.fctr,resY.min,nImgs.cut.fctr:nImgs.cut.fctr,nImgs.cut.fctr:resX.mad.log1p,nImgs.cut.fctr:nImgs.log1p,nImgs.cut.fctr:resY.mean.log1p,nImgs.cut.fctr:resY.mean.root2,nImgs.cut.fctr:resXY.mad,nImgs.cut.fctr:resXY.mad.nexp,nImgs.cut.fctr:resX.mean,nImgs.cut.fctr:resX.mean.nexp,nImgs.cut.fctr:resX.min,nImgs.cut.fctr:resXY.min,nImgs.cut.fctr:resY.min
## Low.cor.X##rcv#glmnet nImgs.cut.fctr,.pos,resX.mad.log1p,resXY.mad.nexp,resY.mean.log1p,resY.mad.nexp,resY.mad,resXY.max.log1p,resXY.max.root2,resXY.max,resY.mad.root2,resY.mad.log1p,nImgs.nexp,resXY.mean.log1p,resXY.mean.root2,.rnorm,resXY.mean,resXY.mad,nImgs,resX.mean,resX.min.nexp,resX.mean.nexp,resY.min.nexp,resX.min,resY.min
## All.X##rcv#glmnet nImgs.cut.fctr,nImgs.log1p,.pos,resX.mad.log1p,resX.mad.root2,resX.mad,resXY.mad.nexp,nImgs.root2,resY.mean.log1p,resY.mean.root2,resY.mean,resY.mad.nexp,resY.mad,resXY.max.log1p,resXY.max.root2,resXY.max,resY.mad.root2,resY.mad.log1p,nImgs.nexp,resXY.mean.log1p,resXY.mean.root2,.rnorm,resXY.mean,resXY.mad.root2,resXY.mad,resX.mad.nexp,resXY.mad.log1p,nImgs,resX.mean.log1p,resX.mean.root2,resX.mean,resX.min.nexp,resX.mean.nexp,resY.mean.nexp,resY.min.nexp,resX.min.log1p,resX.min.root2,resX.min,resXY.min.log1p,resXY.min.root2,resY.min.log1p,resY.min.root2,resXY.min,resY.min
## All.X##rcv#glm nImgs.cut.fctr,nImgs.log1p,.pos,resX.mad.log1p,resX.mad.root2,resX.mad,resXY.mad.nexp,nImgs.root2,resY.mean.log1p,resY.mean.root2,resY.mean,resY.mad.nexp,resY.mad,resXY.max.log1p,resXY.max.root2,resXY.max,resY.mad.root2,resY.mad.log1p,nImgs.nexp,resXY.mean.log1p,resXY.mean.root2,.rnorm,resXY.mean,resXY.mad.root2,resXY.mad,resX.mad.nexp,resXY.mad.log1p,nImgs,resX.mean.log1p,resX.mean.root2,resX.mean,resX.min.nexp,resX.mean.nexp,resY.mean.nexp,resY.min.nexp,resX.min.log1p,resX.min.root2,resX.min,resXY.min.log1p,resXY.min.root2,resY.min.log1p,resY.min.root2,resXY.min,resY.min
## max.nTuningRuns min.elapsedtime.everything
## MFO###myMFO_classfr 0 0.422
## Random###myrandom_classfr 0 0.343
## Max.cor.Y.rcv.1X1###glmnet 0 0.780
## Max.cor.Y##rcv#rpart 5 1.690
## Interact.High.cor.Y##rcv#glmnet 25 22.427
## Low.cor.X##rcv#glmnet 25 6.475
## All.X##rcv#glmnet 25 11.043
## All.X##rcv#glm 1 1.488
## min.elapsedtime.final max.AUCpROC.fit
## MFO###myMFO_classfr 0.002 0.5000000
## Random###myrandom_classfr 0.002 0.4980239
## Max.cor.Y.rcv.1X1###glmnet 0.017 0.5304143
## Max.cor.Y##rcv#rpart 0.016 0.5818446
## Interact.High.cor.Y##rcv#glmnet 1.069 0.6058048
## Low.cor.X##rcv#glmnet 0.633 0.5638247
## All.X##rcv#glmnet 1.193 0.5678247
## All.X##rcv#glm 0.073 0.5797928
## max.Sens.fit max.Spec.fit max.AUCROCR.fit
## MFO###myMFO_classfr 0.000 1.0000000 0.5000000
## Random###myrandom_classfr 0.508 0.4880478 0.4870120
## Max.cor.Y.rcv.1X1###glmnet 0.268 0.7928287 0.5392470
## Max.cor.Y##rcv#rpart 0.586 0.5776892 0.5938127
## Interact.High.cor.Y##rcv#glmnet 0.614 0.5976096 0.6427490
## Low.cor.X##rcv#glmnet 0.540 0.5876494 0.5929761
## All.X##rcv#glmnet 0.548 0.5876494 0.6177371
## All.X##rcv#glm 0.556 0.6035857 0.6320319
## opt.prob.threshold.fit max.f.score.fit
## MFO###myMFO_classfr 0.4 0.6675532
## Random###myrandom_classfr 0.4 0.6675532
## Max.cor.Y.rcv.1X1###glmnet 0.4 0.6675532
## Max.cor.Y##rcv#rpart 0.3 0.6698177
## Interact.High.cor.Y##rcv#glmnet 0.3 0.6716519
## Low.cor.X##rcv#glmnet 0.3 0.6724832
## All.X##rcv#glmnet 0.3 0.6784741
## All.X##rcv#glm 0.3 0.6781530
## max.Accuracy.fit max.AccuracyLower.fit
## MFO###myMFO_classfr 0.5009980 0.4695761
## Random###myrandom_classfr 0.5009980 0.4695761
## Max.cor.Y.rcv.1X1###glmnet 0.5009980 0.4695761
## Max.cor.Y##rcv#rpart 0.5255967 0.4805306
## Interact.High.cor.Y##rcv#glmnet 0.5448976 0.4865121
## Low.cor.X##rcv#glmnet 0.5265996 0.4815272
## All.X##rcv#glmnet 0.5299353 0.4974896
## All.X##rcv#glm 0.5236096 0.5024844
## max.AccuracyUpper.fit max.Kappa.fit
## MFO###myMFO_classfr 0.5324140 0.00000000
## Random###myrandom_classfr 0.5324140 0.00000000
## Max.cor.Y.rcv.1X1###glmnet 0.5324140 0.00000000
## Max.cor.Y##rcv#rpart 0.5433509 0.05146235
## Interact.High.cor.Y##rcv#glmnet 0.5493102 0.08975773
## Low.cor.X##rcv#glmnet 0.5443444 0.05311189
## All.X##rcv#glmnet 0.5602241 0.05971262
## All.X##rcv#glm 0.5651800 0.04708344
## max.AUCpROC.OOB max.Sens.OOB max.Spec.OOB
## MFO###myMFO_classfr 0.5000000 0.0000000 1.0000000
## Random###myrandom_classfr 0.5059679 0.4949698 0.5169661
## Max.cor.Y.rcv.1X1###glmnet 0.5300104 0.2676056 0.7924152
## Max.cor.Y##rcv#rpart 0.5039318 0.4849095 0.5229541
## Interact.High.cor.Y##rcv#glmnet 0.5040041 0.5030181 0.5049900
## Low.cor.X##rcv#glmnet 0.4978875 0.4708249 0.5249501
## All.X##rcv#glmnet 0.5108937 0.4788732 0.5429142
## All.X##rcv#glm 0.5238678 0.4788732 0.5688623
## max.AUCROCR.OOB opt.prob.threshold.OOB
## MFO###myMFO_classfr 0.5000000 0.4
## Random###myrandom_classfr 0.4969618 0.4
## Max.cor.Y.rcv.1X1###glmnet 0.5290506 0.4
## Max.cor.Y##rcv#rpart 0.4985602 0.2
## Interact.High.cor.Y##rcv#glmnet 0.5140745 0.2
## Low.cor.X##rcv#glmnet 0.5012952 0.0
## All.X##rcv#glmnet 0.5204922 0.0
## All.X##rcv#glm 0.5202713 0.0
## max.f.score.OOB max.Accuracy.OOB
## MFO###myMFO_classfr 0.6684456 0.502004
## Random###myrandom_classfr 0.6684456 0.502004
## Max.cor.Y.rcv.1X1###glmnet 0.6684456 0.502004
## Max.cor.Y##rcv#rpart 0.6684456 0.502004
## Interact.High.cor.Y##rcv#glmnet 0.6697987 0.507014
## Low.cor.X##rcv#glmnet 0.6684456 0.502004
## All.X##rcv#glmnet 0.6684456 0.502004
## All.X##rcv#glm 0.6684456 0.502004
## max.AccuracyLower.OOB
## MFO###myMFO_classfr 0.4705156
## Random###myrandom_classfr 0.4705156
## Max.cor.Y.rcv.1X1###glmnet 0.4705156
## Max.cor.Y##rcv#rpart 0.4705156
## Interact.High.cor.Y##rcv#glmnet 0.4755136
## Low.cor.X##rcv#glmnet 0.4705156
## All.X##rcv#glmnet 0.4705156
## All.X##rcv#glm 0.4705156
## max.AccuracyUpper.OOB max.Kappa.OOB
## MFO###myMFO_classfr 0.5334806 0.00000000
## Random###myrandom_classfr 0.5334806 0.00000000
## Max.cor.Y.rcv.1X1###glmnet 0.5334806 0.00000000
## Max.cor.Y##rcv#rpart 0.5334806 0.00000000
## Interact.High.cor.Y##rcv#glmnet 0.5384730 0.01013221
## Low.cor.X##rcv#glmnet 0.5334806 0.00000000
## All.X##rcv#glmnet 0.5334806 0.00000000
## All.X##rcv#glm 0.5334806 0.00000000
## max.AccuracySD.fit max.KappaSD.fit
## MFO###myMFO_classfr NA NA
## Random###myrandom_classfr NA NA
## Max.cor.Y.rcv.1X1###glmnet NA NA
## Max.cor.Y##rcv#rpart 0.02756072 0.05485113
## Interact.High.cor.Y##rcv#glmnet 0.02621502 0.05242679
## Low.cor.X##rcv#glmnet 0.01957750 0.03917575
## All.X##rcv#glmnet 0.02333108 0.04665017
## All.X##rcv#glm 0.01780093 0.03558361rm(ret_lst)
fit.models_1_chunk_df <-
myadd_chunk(fit.models_1_chunk_df, "fit.models_1_end", major.inc = FALSE,
label.minor = "teardown")## label step_major step_minor label_minor bgn end
## 5 fit.models_1_preProc 1 4 preProc 213.742 213.831
## 6 fit.models_1_end 1 5 teardown 213.831 NA
## elapsed
## 5 0.089
## 6 NAglb_chunks_df <- myadd_chunk(glb_chunks_df, "fit.models", major.inc = FALSE)## label step_major step_minor label_minor bgn end elapsed
## 17 fit.models 8 1 1 186.824 213.841 27.017
## 18 fit.models 8 2 2 213.842 NA NAfit.models_2_chunk_df <-
myadd_chunk(NULL, "fit.models_2_bgn", label.minor = "setup")## label step_major step_minor label_minor bgn end elapsed
## 1 fit.models_2_bgn 1 0 setup 217.229 NA NAplt_models_df <- glb_models_df[, -grep("SD|Upper|Lower", names(glb_models_df))]
for (var in grep("^min.", names(plt_models_df), value=TRUE)) {
plt_models_df[, sub("min.", "inv.", var)] <-
#ifelse(all(is.na(tmp <- plt_models_df[, var])), NA, 1.0 / tmp)
1.0 / plt_models_df[, var]
plt_models_df <- plt_models_df[ , -grep(var, names(plt_models_df))]
}
print(plt_models_df)## id
## MFO###myMFO_classfr MFO###myMFO_classfr
## Random###myrandom_classfr Random###myrandom_classfr
## Max.cor.Y.rcv.1X1###glmnet Max.cor.Y.rcv.1X1###glmnet
## Max.cor.Y##rcv#rpart Max.cor.Y##rcv#rpart
## Interact.High.cor.Y##rcv#glmnet Interact.High.cor.Y##rcv#glmnet
## Low.cor.X##rcv#glmnet Low.cor.X##rcv#glmnet
## All.X##rcv#glmnet All.X##rcv#glmnet
## All.X##rcv#glm All.X##rcv#glm
## feats
## MFO###myMFO_classfr .rnorm
## Random###myrandom_classfr .rnorm
## Max.cor.Y.rcv.1X1###glmnet nImgs.cut.fctr,resY.min
## Max.cor.Y##rcv#rpart nImgs.cut.fctr,resY.min
## Interact.High.cor.Y##rcv#glmnet nImgs.cut.fctr,resY.min,nImgs.cut.fctr:nImgs.cut.fctr,nImgs.cut.fctr:resX.mad.log1p,nImgs.cut.fctr:nImgs.log1p,nImgs.cut.fctr:resY.mean.log1p,nImgs.cut.fctr:resY.mean.root2,nImgs.cut.fctr:resXY.mad,nImgs.cut.fctr:resXY.mad.nexp,nImgs.cut.fctr:resX.mean,nImgs.cut.fctr:resX.mean.nexp,nImgs.cut.fctr:resX.min,nImgs.cut.fctr:resXY.min,nImgs.cut.fctr:resY.min
## Low.cor.X##rcv#glmnet nImgs.cut.fctr,.pos,resX.mad.log1p,resXY.mad.nexp,resY.mean.log1p,resY.mad.nexp,resY.mad,resXY.max.log1p,resXY.max.root2,resXY.max,resY.mad.root2,resY.mad.log1p,nImgs.nexp,resXY.mean.log1p,resXY.mean.root2,.rnorm,resXY.mean,resXY.mad,nImgs,resX.mean,resX.min.nexp,resX.mean.nexp,resY.min.nexp,resX.min,resY.min
## All.X##rcv#glmnet nImgs.cut.fctr,nImgs.log1p,.pos,resX.mad.log1p,resX.mad.root2,resX.mad,resXY.mad.nexp,nImgs.root2,resY.mean.log1p,resY.mean.root2,resY.mean,resY.mad.nexp,resY.mad,resXY.max.log1p,resXY.max.root2,resXY.max,resY.mad.root2,resY.mad.log1p,nImgs.nexp,resXY.mean.log1p,resXY.mean.root2,.rnorm,resXY.mean,resXY.mad.root2,resXY.mad,resX.mad.nexp,resXY.mad.log1p,nImgs,resX.mean.log1p,resX.mean.root2,resX.mean,resX.min.nexp,resX.mean.nexp,resY.mean.nexp,resY.min.nexp,resX.min.log1p,resX.min.root2,resX.min,resXY.min.log1p,resXY.min.root2,resY.min.log1p,resY.min.root2,resXY.min,resY.min
## All.X##rcv#glm nImgs.cut.fctr,nImgs.log1p,.pos,resX.mad.log1p,resX.mad.root2,resX.mad,resXY.mad.nexp,nImgs.root2,resY.mean.log1p,resY.mean.root2,resY.mean,resY.mad.nexp,resY.mad,resXY.max.log1p,resXY.max.root2,resXY.max,resY.mad.root2,resY.mad.log1p,nImgs.nexp,resXY.mean.log1p,resXY.mean.root2,.rnorm,resXY.mean,resXY.mad.root2,resXY.mad,resX.mad.nexp,resXY.mad.log1p,nImgs,resX.mean.log1p,resX.mean.root2,resX.mean,resX.min.nexp,resX.mean.nexp,resY.mean.nexp,resY.min.nexp,resX.min.log1p,resX.min.root2,resX.min,resXY.min.log1p,resXY.min.root2,resY.min.log1p,resY.min.root2,resXY.min,resY.min
## max.nTuningRuns max.AUCpROC.fit
## MFO###myMFO_classfr 0 0.5000000
## Random###myrandom_classfr 0 0.4980239
## Max.cor.Y.rcv.1X1###glmnet 0 0.5304143
## Max.cor.Y##rcv#rpart 5 0.5818446
## Interact.High.cor.Y##rcv#glmnet 25 0.6058048
## Low.cor.X##rcv#glmnet 25 0.5638247
## All.X##rcv#glmnet 25 0.5678247
## All.X##rcv#glm 1 0.5797928
## max.Sens.fit max.Spec.fit max.AUCROCR.fit
## MFO###myMFO_classfr 0.000 1.0000000 0.5000000
## Random###myrandom_classfr 0.508 0.4880478 0.4870120
## Max.cor.Y.rcv.1X1###glmnet 0.268 0.7928287 0.5392470
## Max.cor.Y##rcv#rpart 0.586 0.5776892 0.5938127
## Interact.High.cor.Y##rcv#glmnet 0.614 0.5976096 0.6427490
## Low.cor.X##rcv#glmnet 0.540 0.5876494 0.5929761
## All.X##rcv#glmnet 0.548 0.5876494 0.6177371
## All.X##rcv#glm 0.556 0.6035857 0.6320319
## opt.prob.threshold.fit max.f.score.fit
## MFO###myMFO_classfr 0.4 0.6675532
## Random###myrandom_classfr 0.4 0.6675532
## Max.cor.Y.rcv.1X1###glmnet 0.4 0.6675532
## Max.cor.Y##rcv#rpart 0.3 0.6698177
## Interact.High.cor.Y##rcv#glmnet 0.3 0.6716519
## Low.cor.X##rcv#glmnet 0.3 0.6724832
## All.X##rcv#glmnet 0.3 0.6784741
## All.X##rcv#glm 0.3 0.6781530
## max.Accuracy.fit max.Kappa.fit
## MFO###myMFO_classfr 0.5009980 0.00000000
## Random###myrandom_classfr 0.5009980 0.00000000
## Max.cor.Y.rcv.1X1###glmnet 0.5009980 0.00000000
## Max.cor.Y##rcv#rpart 0.5255967 0.05146235
## Interact.High.cor.Y##rcv#glmnet 0.5448976 0.08975773
## Low.cor.X##rcv#glmnet 0.5265996 0.05311189
## All.X##rcv#glmnet 0.5299353 0.05971262
## All.X##rcv#glm 0.5236096 0.04708344
## max.AUCpROC.OOB max.Sens.OOB max.Spec.OOB
## MFO###myMFO_classfr 0.5000000 0.0000000 1.0000000
## Random###myrandom_classfr 0.5059679 0.4949698 0.5169661
## Max.cor.Y.rcv.1X1###glmnet 0.5300104 0.2676056 0.7924152
## Max.cor.Y##rcv#rpart 0.5039318 0.4849095 0.5229541
## Interact.High.cor.Y##rcv#glmnet 0.5040041 0.5030181 0.5049900
## Low.cor.X##rcv#glmnet 0.4978875 0.4708249 0.5249501
## All.X##rcv#glmnet 0.5108937 0.4788732 0.5429142
## All.X##rcv#glm 0.5238678 0.4788732 0.5688623
## max.AUCROCR.OOB opt.prob.threshold.OOB
## MFO###myMFO_classfr 0.5000000 0.4
## Random###myrandom_classfr 0.4969618 0.4
## Max.cor.Y.rcv.1X1###glmnet 0.5290506 0.4
## Max.cor.Y##rcv#rpart 0.4985602 0.2
## Interact.High.cor.Y##rcv#glmnet 0.5140745 0.2
## Low.cor.X##rcv#glmnet 0.5012952 0.0
## All.X##rcv#glmnet 0.5204922 0.0
## All.X##rcv#glm 0.5202713 0.0
## max.f.score.OOB max.Accuracy.OOB
## MFO###myMFO_classfr 0.6684456 0.502004
## Random###myrandom_classfr 0.6684456 0.502004
## Max.cor.Y.rcv.1X1###glmnet 0.6684456 0.502004
## Max.cor.Y##rcv#rpart 0.6684456 0.502004
## Interact.High.cor.Y##rcv#glmnet 0.6697987 0.507014
## Low.cor.X##rcv#glmnet 0.6684456 0.502004
## All.X##rcv#glmnet 0.6684456 0.502004
## All.X##rcv#glm 0.6684456 0.502004
## max.Kappa.OOB inv.elapsedtime.everything
## MFO###myMFO_classfr 0.00000000 2.36966825
## Random###myrandom_classfr 0.00000000 2.91545190
## Max.cor.Y.rcv.1X1###glmnet 0.00000000 1.28205128
## Max.cor.Y##rcv#rpart 0.00000000 0.59171598
## Interact.High.cor.Y##rcv#glmnet 0.01013221 0.04458911
## Low.cor.X##rcv#glmnet 0.00000000 0.15444015
## All.X##rcv#glmnet 0.00000000 0.09055510
## All.X##rcv#glm 0.00000000 0.67204301
## inv.elapsedtime.final
## MFO###myMFO_classfr 500.0000000
## Random###myrandom_classfr 500.0000000
## Max.cor.Y.rcv.1X1###glmnet 58.8235294
## Max.cor.Y##rcv#rpart 62.5000000
## Interact.High.cor.Y##rcv#glmnet 0.9354537
## Low.cor.X##rcv#glmnet 1.5797788
## All.X##rcv#glmnet 0.8382230
## All.X##rcv#glm 13.6986301# print(myplot_radar(radar_inp_df=plt_models_df))
# print(myplot_radar(radar_inp_df=subset(plt_models_df,
# !(mdl_id %in% grep("random|MFO", plt_models_df$id, value=TRUE)))))
# Compute CI for <metric>SD
glb_models_df <- mutate(glb_models_df,
max.df = ifelse(max.nTuningRuns > 1, max.nTuningRuns - 1, NA),
min.sd2ci.scaler = ifelse(is.na(max.df), NA, qt(0.975, max.df)))
for (var in grep("SD", names(glb_models_df), value=TRUE)) {
# Does CI alredy exist ?
var_components <- unlist(strsplit(var, "SD"))
varActul <- paste0(var_components[1], var_components[2])
varUpper <- paste0(var_components[1], "Upper", var_components[2])
varLower <- paste0(var_components[1], "Lower", var_components[2])
if (varUpper %in% names(glb_models_df)) {
warning(varUpper, " already exists in glb_models_df")
# Assuming Lower also exists
next
}
print(sprintf("var:%s", var))
# CI is dependent on sample size in t distribution; df=n-1
glb_models_df[, varUpper] <- glb_models_df[, varActul] +
glb_models_df[, "min.sd2ci.scaler"] * glb_models_df[, var]
glb_models_df[, varLower] <- glb_models_df[, varActul] -
glb_models_df[, "min.sd2ci.scaler"] * glb_models_df[, var]
}## Warning: max.AccuracyUpper.fit already exists in glb_models_df## [1] "var:max.KappaSD.fit"# Plot metrics with CI
plt_models_df <- glb_models_df[, "id", FALSE]
pltCI_models_df <- glb_models_df[, "id", FALSE]
for (var in grep("Upper", names(glb_models_df), value=TRUE)) {
var_components <- unlist(strsplit(var, "Upper"))
col_name <- unlist(paste(var_components, collapse=""))
plt_models_df[, col_name] <- glb_models_df[, col_name]
for (name in paste0(var_components[1], c("Upper", "Lower"), var_components[2]))
pltCI_models_df[, name] <- glb_models_df[, name]
}
build_statsCI_data <- function(plt_models_df) {
mltd_models_df <- melt(plt_models_df, id.vars="id")
mltd_models_df$data <- sapply(1:nrow(mltd_models_df),
function(row_ix) tail(unlist(strsplit(as.character(
mltd_models_df[row_ix, "variable"]), "[.]")), 1))
mltd_models_df$label <- sapply(1:nrow(mltd_models_df),
function(row_ix) head(unlist(strsplit(as.character(
mltd_models_df[row_ix, "variable"]),
paste0(".", mltd_models_df[row_ix, "data"]))), 1))
#print(mltd_models_df)
return(mltd_models_df)
}
mltd_models_df <- build_statsCI_data(plt_models_df)
mltdCI_models_df <- melt(pltCI_models_df, id.vars="id")
for (row_ix in 1:nrow(mltdCI_models_df)) {
for (type in c("Upper", "Lower")) {
if (length(var_components <- unlist(strsplit(
as.character(mltdCI_models_df[row_ix, "variable"]), type))) > 1) {
#print(sprintf("row_ix:%d; type:%s; ", row_ix, type))
mltdCI_models_df[row_ix, "label"] <- var_components[1]
mltdCI_models_df[row_ix, "data"] <-
unlist(strsplit(var_components[2], "[.]"))[2]
mltdCI_models_df[row_ix, "type"] <- type
break
}
}
}
wideCI_models_df <- reshape(subset(mltdCI_models_df, select=-variable),
timevar="type",
idvar=setdiff(names(mltdCI_models_df), c("type", "value", "variable")),
direction="wide")
#print(wideCI_models_df)
mrgdCI_models_df <- merge(wideCI_models_df, mltd_models_df, all.x=TRUE)
#print(mrgdCI_models_df)
# Merge stats back in if CIs don't exist
goback_vars <- c()
for (var in unique(mltd_models_df$label)) {
for (type in unique(mltd_models_df$data)) {
var_type <- paste0(var, ".", type)
# if this data is already present, next
if (var_type %in% unique(paste(mltd_models_df$label, mltd_models_df$data,
sep=".")))
next
#print(sprintf("var_type:%s", var_type))
goback_vars <- c(goback_vars, var_type)
}
}
if (length(goback_vars) > 0) {
mltd_goback_df <- build_statsCI_data(glb_models_df[, c("id", goback_vars)])
mltd_models_df <- rbind(mltd_models_df, mltd_goback_df)
}
# mltd_models_df <- merge(mltd_models_df, glb_models_df[, c("id", "model_method")],
# all.x=TRUE)
png(paste0(glbOut$pfx, "models_bar.png"), width=480*3, height=480*2)
#print(gp <- myplot_bar(mltd_models_df, "id", "value", colorcol_name="model_method") +
print(gp <- myplot_bar(df=mltd_models_df, xcol_name="id", ycol_names="value") +
geom_errorbar(data=mrgdCI_models_df,
mapping=aes(x=mdl_id, ymax=value.Upper, ymin=value.Lower), width=0.5) +
facet_grid(label ~ data, scales="free") +
theme(axis.text.x = element_text(angle = 90,vjust = 0.5)))## Warning: Removed 4 rows containing missing values (geom_errorbar).dev.off()## quartz_off_screen
## 2print(gp)## Warning: Removed 4 rows containing missing values (geom_errorbar).
dsp_models_cols <- c("id",
glbMdlMetricsEval[glbMdlMetricsEval %in% names(glb_models_df)],
grep("opt.", names(glb_models_df), fixed = TRUE, value = TRUE))
# if (glb_is_classification && glb_is_binomial)
# dsp_models_cols <- c(dsp_models_cols, "opt.prob.threshold.OOB")
print(dsp_models_df <- orderBy(get_model_sel_frmla(), glb_models_df)[, dsp_models_cols])## id max.Accuracy.OOB max.AUCROCR.OOB
## 5 Interact.High.cor.Y##rcv#glmnet 0.507014 0.5140745
## 3 Max.cor.Y.rcv.1X1###glmnet 0.502004 0.5290506
## 7 All.X##rcv#glmnet 0.502004 0.5204922
## 8 All.X##rcv#glm 0.502004 0.5202713
## 6 Low.cor.X##rcv#glmnet 0.502004 0.5012952
## 1 MFO###myMFO_classfr 0.502004 0.5000000
## 4 Max.cor.Y##rcv#rpart 0.502004 0.4985602
## 2 Random###myrandom_classfr 0.502004 0.4969618
## max.AUCpROC.OOB max.Accuracy.fit opt.prob.threshold.fit
## 5 0.5040041 0.5448976 0.3
## 3 0.5300104 0.5009980 0.4
## 7 0.5108937 0.5299353 0.3
## 8 0.5238678 0.5236096 0.3
## 6 0.4978875 0.5265996 0.3
## 1 0.5000000 0.5009980 0.4
## 4 0.5039318 0.5255967 0.3
## 2 0.5059679 0.5009980 0.4
## opt.prob.threshold.OOB
## 5 0.2
## 3 0.4
## 7 0.0
## 8 0.0
## 6 0.0
## 1 0.4
## 4 0.2
## 2 0.4# print(myplot_radar(radar_inp_df = dsp_models_df))
print("Metrics used for model selection:"); print(get_model_sel_frmla())## [1] "Metrics used for model selection:"## ~-max.Accuracy.OOB - max.AUCROCR.OOB - max.AUCpROC.OOB - max.Accuracy.fit -
## opt.prob.threshold.OOB
## <environment: 0x7fc77ac162d0>print(sprintf("Best model id: %s", dsp_models_df[1, "id"]))## [1] "Best model id: Interact.High.cor.Y##rcv#glmnet"glb_get_predictions <- function(df, mdl_id, rsp_var, prob_threshold_def=NULL, verbose=FALSE) {
mdl <- glb_models_lst[[mdl_id]]
clmnNames <- mygetPredictIds(rsp_var, mdl_id)
predct_var_name <- clmnNames$value
predct_prob_var_name <- clmnNames$prob
predct_accurate_var_name <- clmnNames$is.acc
predct_error_var_name <- clmnNames$err
predct_erabs_var_name <- clmnNames$err.abs
if (glb_is_regression) {
df[, predct_var_name] <- predict(mdl, newdata=df, type="raw")
if (verbose) print(myplot_scatter(df, glb_rsp_var, predct_var_name) +
facet_wrap(reformulate(glbFeatsCategory), scales = "free") +
stat_smooth(method="glm"))
df[, predct_error_var_name] <- df[, predct_var_name] - df[, glb_rsp_var]
if (verbose) print(myplot_scatter(df, predct_var_name, predct_error_var_name) +
#facet_wrap(reformulate(glbFeatsCategory), scales = "free") +
stat_smooth(method="auto"))
if (verbose) print(myplot_scatter(df, glb_rsp_var, predct_error_var_name) +
#facet_wrap(reformulate(glbFeatsCategory), scales = "free") +
stat_smooth(method="glm"))
df[, predct_erabs_var_name] <- abs(df[, predct_error_var_name])
if (verbose) print(head(orderBy(reformulate(c("-", predct_erabs_var_name)), df)))
df[, predct_accurate_var_name] <- (df[, glb_rsp_var] == df[, predct_var_name])
}
if (glb_is_classification && glb_is_binomial) {
prob_threshold <- glb_models_df[glb_models_df$id == mdl_id,
"opt.prob.threshold.OOB"]
if (is.null(prob_threshold) || is.na(prob_threshold)) {
warning("Using default probability threshold: ", prob_threshold_def)
if (is.null(prob_threshold <- prob_threshold_def))
stop("Default probability threshold is NULL")
}
df[, predct_prob_var_name] <- predict(mdl, newdata = df, type = "prob")[, 2]
df[, predct_var_name] <-
factor(levels(df[, glb_rsp_var])[
(df[, predct_prob_var_name] >=
prob_threshold) * 1 + 1], levels(df[, glb_rsp_var]))
# if (verbose) print(myplot_scatter(df, glb_rsp_var, predct_var_name) +
# facet_wrap(reformulate(glbFeatsCategory), scales = "free") +
# stat_smooth(method="glm"))
df[, predct_error_var_name] <- df[, predct_var_name] != df[, glb_rsp_var]
# if (verbose) print(myplot_scatter(df, predct_var_name, predct_error_var_name) +
# #facet_wrap(reformulate(glbFeatsCategory), scales = "free") +
# stat_smooth(method="auto"))
# if (verbose) print(myplot_scatter(df, glb_rsp_var, predct_error_var_name) +
# #facet_wrap(reformulate(glbFeatsCategory), scales = "free") +
# stat_smooth(method="glm"))
# if prediction is a TP (true +ve), measure distance from 1.0
tp <- which((df[, predct_var_name] == df[, glb_rsp_var]) &
(df[, predct_var_name] == levels(df[, glb_rsp_var])[2]))
df[tp, predct_erabs_var_name] <- abs(1 - df[tp, predct_prob_var_name])
#rowIx <- which.max(df[tp, predct_erabs_var_name]); df[tp, c(glbFeatsId, glb_rsp_var, predct_var_name, predct_prob_var_name, predct_erabs_var_name)][rowIx, ]
# if prediction is a TN (true -ve), measure distance from 0.0
tn <- which((df[, predct_var_name] == df[, glb_rsp_var]) &
(df[, predct_var_name] == levels(df[, glb_rsp_var])[1]))
df[tn, predct_erabs_var_name] <- abs(0 - df[tn, predct_prob_var_name])
#rowIx <- which.max(df[tn, predct_erabs_var_name]); df[tn, c(glbFeatsId, glb_rsp_var, predct_var_name, predct_prob_var_name, predct_erabs_var_name)][rowIx, ]
# if prediction is a FP (flse +ve), measure distance from 0.0
fp <- which((df[, predct_var_name] != df[, glb_rsp_var]) &
(df[, predct_var_name] == levels(df[, glb_rsp_var])[2]))
df[fp, predct_erabs_var_name] <- abs(0 - df[fp, predct_prob_var_name])
#rowIx <- which.max(df[fp, predct_erabs_var_name]); df[fp, c(glbFeatsId, glb_rsp_var, predct_var_name, predct_prob_var_name, predct_erabs_var_name)][rowIx, ]
# if prediction is a FN (flse -ve), measure distance from 1.0
fn <- which((df[, predct_var_name] != df[, glb_rsp_var]) &
(df[, predct_var_name] == levels(df[, glb_rsp_var])[1]))
df[fn, predct_erabs_var_name] <- abs(1 - df[fn, predct_prob_var_name])
#rowIx <- which.max(df[fn, predct_erabs_var_name]); df[fn, c(glbFeatsId, glb_rsp_var, predct_var_name, predct_prob_var_name, predct_erabs_var_name)][rowIx, ]
if (verbose) print(head(orderBy(reformulate(c("-", predct_erabs_var_name)), df)))
df[, predct_accurate_var_name] <- (df[, glb_rsp_var] == df[, predct_var_name])
}
if (glb_is_classification && !glb_is_binomial) {
df[, predct_var_name] <- predict(mdl, newdata = df, type = "raw")
probCls <- predict(mdl, newdata = df, type = "prob")
df[, predct_prob_var_name] <- NA
for (cls in names(probCls)) {
mask <- (df[, predct_var_name] == cls)
df[mask, predct_prob_var_name] <- probCls[mask, cls]
}
if (verbose) print(myplot_histogram(df, predct_prob_var_name,
fill_col_name = predct_var_name))
if (verbose) print(myplot_histogram(df, predct_prob_var_name,
facet_frmla = paste0("~", glb_rsp_var)))
df[, predct_error_var_name] <- df[, predct_var_name] != df[, glb_rsp_var]
# if prediction is erroneous, measure predicted class prob from actual class prob
df[, predct_erabs_var_name] <- 0
for (cls in names(probCls)) {
mask <- (df[, glb_rsp_var] == cls) & (df[, predct_error_var_name])
df[mask, predct_erabs_var_name] <- probCls[mask, cls]
}
df[, predct_accurate_var_name] <- (df[, glb_rsp_var] == df[, predct_var_name])
}
return(df)
}
#stop(here"); glb2Sav(); glbObsAll <- savObsAll; glbObsTrn <- savObsTrn; glbObsFit <- savObsFit; glbObsOOB <- savObsOOB; sav_models_df <- glb_models_df; glb_models_df <- sav_models_df; glb_featsimp_df <- sav_featsimp_df
myget_category_stats <- function(obs_df, mdl_id, label) {
require(dplyr)
require(lazyeval)
predct_var_name <- mygetPredictIds(glb_rsp_var, mdl_id)$value
predct_error_var_name <- mygetPredictIds(glb_rsp_var, mdl_id)$err.abs
if (!predct_var_name %in% names(obs_df))
obs_df <- glb_get_predictions(obs_df, mdl_id, glb_rsp_var)
tmp_obs_df <- obs_df[, c(glbFeatsCategory, glb_rsp_var,
predct_var_name, predct_error_var_name)]
# tmp_obs_df <- obs_df %>%
# dplyr::select_(glbFeatsCategory, glb_rsp_var, predct_var_name, predct_error_var_name)
#dplyr::rename(startprice.log10.predict.RFE.X.glmnet.err=error_abs_OOB)
names(tmp_obs_df)[length(names(tmp_obs_df))] <- paste0("err.abs.", label)
ret_ctgry_df <- tmp_obs_df %>%
dplyr::group_by_(glbFeatsCategory) %>%
dplyr::summarise_(#interp(~sum(abs(var)), var=as.name(glb_rsp_var)),
interp(~sum(var), var=as.name(paste0("err.abs.", label))),
interp(~mean(var), var=as.name(paste0("err.abs.", label))),
interp(~n()))
names(ret_ctgry_df) <- c(glbFeatsCategory,
#paste0(glb_rsp_var, ".abs.", label, ".sum"),
paste0("err.abs.", label, ".sum"),
paste0("err.abs.", label, ".mean"),
paste0(".n.", label))
ret_ctgry_df <- dplyr::ungroup(ret_ctgry_df)
#colSums(ret_ctgry_df[, -grep(glbFeatsCategory, names(ret_ctgry_df))])
return(ret_ctgry_df)
}
#print(colSums((ctgry_df <- myget_category_stats(obs_df=glbObsFit, mdl_id="", label="fit"))[, -grep(glbFeatsCategory, names(ctgry_df))]))
if (!is.null(glb_mdl_ensemble)) {
fit.models_2_chunk_df <- myadd_chunk(fit.models_2_chunk_df,
paste0("fit.models_2_", mdl_id_pfx), major.inc = TRUE,
label.minor = "ensemble")
mdl_id_pfx <- "Ensemble"
if (#(glb_is_regression) |
((glb_is_classification) & (!glb_is_binomial)))
stop("Ensemble models not implemented yet for multinomial classification")
mygetEnsembleAutoMdlIds <- function() {
tmp_models_df <- orderBy(get_model_sel_frmla(), glb_models_df)
row.names(tmp_models_df) <- tmp_models_df$id
mdl_threshold_pos <-
min(which(grepl("MFO|Random|Baseline", tmp_models_df$id))) - 1
mdlIds <- tmp_models_df$id[1:mdl_threshold_pos]
return(mdlIds[!grepl("Ensemble", mdlIds)])
}
if (glb_mdl_ensemble == "auto") {
glb_mdl_ensemble <- mygetEnsembleAutoMdlIds()
mdl_id_pfx <- paste0(mdl_id_pfx, ".auto")
} else if (grepl("^%<d-%", glb_mdl_ensemble)) {
glb_mdl_ensemble <- eval(parse(text =
str_trim(unlist(strsplit(glb_mdl_ensemble, "%<d-%"))[2])))
}
for (mdl_id in glb_mdl_ensemble) {
if (!(mdl_id %in% names(glb_models_lst))) {
warning("Model ", mdl_id, " in glb_model_ensemble not found !")
next
}
glbObsFit <- glb_get_predictions(df = glbObsFit, mdl_id, glb_rsp_var)
glbObsOOB <- glb_get_predictions(df = glbObsOOB, mdl_id, glb_rsp_var)
}
#mdl_id_pfx <- "Ensemble.RFE"; mdlId <- paste0(mdl_id_pfx, ".glmnet")
#glb_mdl_ensemble <- gsub(mygetPredictIds$value, "", grep("RFE\\.X\\.(?!Interact)", row.names(glb_featsimp_df), perl = TRUE, value = TRUE), fixed = TRUE)
#varImp(glb_models_lst[[mdlId]])
#cor_df <- data.frame(cor=cor(glbObsFit[, glb_rsp_var], glbObsFit[, paste(mygetPredictIds$value, glb_mdl_ensemble)], use="pairwise.complete.obs"))
#glbObsFit <- glb_get_predictions(df=glbObsFit, "Ensemble.glmnet", glb_rsp_var);print(colSums((ctgry_df <- myget_category_stats(obs_df=glbObsFit, mdl_id="Ensemble.glmnet", label="fit"))[, -grep(glbFeatsCategory, names(ctgry_df))]))
### bid0_sp
# Better than MFO; models.n=28; min.RMSE.fit=0.0521233; err.abs.fit.sum=7.3631895
# old: Top x from auto; models.n= 5; min.RMSE.fit=0.06311047; err.abs.fit.sum=9.5937080
# RFE only ; models.n=16; min.RMSE.fit=0.05148588; err.abs.fit.sum=7.2875091
# RFE subset only ;models.n= 5; min.RMSE.fit=0.06040702; err.abs.fit.sum=9.059088
# RFE subset only ;models.n= 9; min.RMSE.fit=0.05933167; err.abs.fit.sum=8.7421288
# RFE subset only ;models.n=15; min.RMSE.fit=0.0584607; err.abs.fit.sum=8.5902066
# RFE subset only ;models.n=17; min.RMSE.fit=0.05496899; err.abs.fit.sum=8.0170431
# RFE subset only ;models.n=18; min.RMSE.fit=0.05441577; err.abs.fit.sum=7.837223
# RFE subset only ;models.n=16; min.RMSE.fit=0.05441577; err.abs.fit.sum=7.837223
### bid0_sp
### bid1_sp
# "auto"; err.abs.fit.sum=76.699774; min.RMSE.fit=0.2186429
# "RFE.X.*"; err.abs.fit.sum=; min.RMSE.fit=0.221114
### bid1_sp
indep_vars <- paste(mygetPredictIds(glb_rsp_var)$value, glb_mdl_ensemble, sep = "")
if (glb_is_classification)
indep_vars <- paste(indep_vars, ".prob", sep = "")
# Some models in glb_mdl_ensemble might not be fitted e.g. RFE.X.Interact
indep_vars <- intersect(indep_vars, names(glbObsFit))
# indep_vars <- grep(mygetPredictIds(glb_rsp_var)$value, names(glbObsFit), fixed=TRUE, value=TRUE)
# if (glb_is_regression)
# indep_vars <- indep_vars[!grepl("(err\\.abs|accurate)$", indep_vars)]
# if (glb_is_classification && glb_is_binomial)
# indep_vars <- grep("prob$", indep_vars, value=TRUE) else
# indep_vars <- indep_vars[!grepl("err$", indep_vars)]
#rfe_fit_ens_results <- myrun_rfe(glbObsFit, indep_vars)
for (method in c("glm", "glmnet")) {
for (trainControlMethod in
c("boot", "boot632", "cv", "repeatedcv"
#, "LOOCV" # tuneLength * nrow(fitDF)
, "LGOCV", "adaptive_cv"
#, "adaptive_boot" #error: adaptive$min should be less than 3
#, "adaptive_LGOCV" #error: adaptive$min should be less than 3
)) {
#sav_models_df <- glb_models_df; all.equal(sav_models_df, glb_models_df)
#glb_models_df <- sav_models_df; print(glb_models_df$id)
if ((method == "glm") && (trainControlMethod != "repeatedcv"))
# glm used only to identify outliers
next
ret_lst <- myfit_mdl(
mdl_specs_lst = myinit_mdl_specs_lst(mdl_specs_lst = list(
id.prefix = paste0(mdl_id_pfx, ".", trainControlMethod),
type = glb_model_type, tune.df = NULL,
trainControl.method = trainControlMethod,
trainControl.number = glb_rcv_n_folds,
trainControl.repeats = glb_rcv_n_repeats,
trainControl.classProbs = glb_is_classification,
trainControl.summaryFunction = glbMdlMetricSummaryFn,
train.metric = glbMdlMetricSummary,
train.maximize = glbMdlMetricMaximize,
train.method = method)),
indep_vars = indep_vars, rsp_var = glb_rsp_var,
fit_df = glbObsFit, OOB_df = glbObsOOB)
}
}
dsp_models_df <- get_dsp_models_df()
}
if (is.null(glb_sel_mdl_id))
glb_sel_mdl_id <- dsp_models_df[1, "id"] else
print(sprintf("User specified selection: %s", glb_sel_mdl_id)) ## [1] "User specified selection: All.X##rcv#glmnet"myprint_mdl(glb_sel_mdl <- glb_models_lst[[glb_sel_mdl_id]])
## Length Class Mode
## a0 100 -none- numeric
## beta 4600 dgCMatrix S4
## df 100 -none- numeric
## dim 2 -none- numeric
## lambda 100 -none- numeric
## dev.ratio 100 -none- numeric
## nulldev 1 -none- numeric
## npasses 1 -none- numeric
## jerr 1 -none- numeric
## offset 1 -none- logical
## classnames 2 -none- character
## call 5 -none- call
## nobs 1 -none- numeric
## lambdaOpt 1 -none- numeric
## xNames 46 -none- character
## problemType 1 -none- character
## tuneValue 2 data.frame list
## obsLevels 2 -none- character
## [1] "min lambda > lambdaOpt:"
## (Intercept) .pos
## -5.322001e+02 1.037459e-04
## .rnorm nImgs
## -6.769060e-02 1.521544e-04
## nImgs.cut.fctr(32,60] nImgs.cut.fctr(60,120]
## 1.309604e-01 6.523010e-02
## nImgs.cut.fctr(120,3e+03] nImgs.log1p
## 3.460511e-01 5.930509e-01
## nImgs.nexp nImgs.root2
## 5.294506e+01 -1.033055e-01
## resX.mad resX.mad.log1p
## -7.462293e-03 3.347949e-01
## resX.mad.nexp resX.mad.root2
## 8.597647e-01 1.847927e-02
## resX.mean resX.mean.log1p
## -2.908604e-02 1.718262e+01
## resX.mean.nexp resX.mean.root2
## 9.900000e+35 -1.044154e-01
## resX.min resX.min.log1p
## 1.111873e-02 -9.111545e+00
## resX.min.nexp resX.min.root2
## -3.246667e+27 7.845843e-01
## resXY.mad resXY.mad.log1p
## 2.701393e-06 1.155758e-01
## resXY.mad.nexp resXY.mad.root2
## 1.030157e+00 -2.916969e-03
## resXY.max resXY.max.log1p
## -1.265518e-04 3.573686e+01
## resXY.max.root2 resXY.mean
## -3.382568e-02 -1.511560e-04
## resXY.mean.log1p resXY.mean.root2
## 1.984867e+01 -6.256731e-03
## resXY.min resXY.min.log1p
## -3.496593e-05 1.785193e+00
## resXY.min.root2 resY.mad
## 8.746218e-03 1.065394e-02
## resY.mad.log1p resY.mad.nexp
## 1.160382e+00 1.082350e+00
## resY.mad.root2 resY.mean
## -5.316544e-01 7.943836e-02
## resY.mean.log1p resY.mean.nexp
## -2.938950e+01 9.900000e+35
## resY.mean.root2 resY.min
## 1.030505e-01 9.792523e-04
## resY.min.log1p resY.min.nexp
## -2.872098e+00 -6.257821e+21
## resY.min.root2
## 3.282024e-01
## [1] "max lambda < lambdaOpt:"
## [1] "Feats mismatch between coefs_left & rght:"
## [1] "(Intercept)" ".pos"
## [3] ".rnorm" "nImgs"
## [5] "nImgs.cut.fctr(32,60]" "nImgs.cut.fctr(60,120]"
## [7] "nImgs.cut.fctr(120,3e+03]" "nImgs.log1p"
## [9] "nImgs.nexp" "nImgs.root2"
## [11] "resX.mad" "resX.mad.log1p"
## [13] "resX.mad.nexp" "resX.mad.root2"
## [15] "resX.mean" "resX.mean.log1p"
## [17] "resX.mean.nexp" "resX.mean.root2"
## [19] "resX.min" "resX.min.log1p"
## [21] "resX.min.nexp" "resX.min.root2"
## [23] "resXY.mad" "resXY.mad.log1p"
## [25] "resXY.mad.nexp" "resXY.mad.root2"
## [27] "resXY.max" "resXY.max.log1p"
## [29] "resXY.max.root2" "resXY.mean"
## [31] "resXY.mean.log1p" "resXY.mean.root2"
## [33] "resXY.min" "resXY.min.log1p"
## [35] "resXY.min.root2" "resY.mad"
## [37] "resY.mad.log1p" "resY.mad.nexp"
## [39] "resY.mad.root2" "resY.mean"
## [41] "resY.mean.log1p" "resY.mean.nexp"
## [43] "resY.mean.root2" "resY.min"
## [45] "resY.min.log1p" "resY.min.nexp"
## [47] "resY.min.root2"## [1] TRUE# From here to save(), this should all be in one function
# these are executed in the same seq twice more:
# fit.data.training & predict.data.new chunks
print(sprintf("%s fit prediction diagnostics:", glb_sel_mdl_id))## [1] "All.X##rcv#glmnet fit prediction diagnostics:"glbObsFit <- glb_get_predictions(df = glbObsFit, mdl_id = glb_sel_mdl_id,
rsp_var = glb_rsp_var)
print(sprintf("%s OOB prediction diagnostics:", glb_sel_mdl_id))## [1] "All.X##rcv#glmnet OOB prediction diagnostics:"glbObsOOB <- glb_get_predictions(df = glbObsOOB, mdl_id = glb_sel_mdl_id,
rsp_var = glb_rsp_var)
print(glb_featsimp_df <- myget_feats_importance(mdl = glb_sel_mdl, featsimp_df = NULL))## All.X..rcv.glmnet.imp imp
## resX.mean.nexp 1.000000e+02 1.000000e+02
## resY.mean.nexp 1.000000e+02 1.000000e+02
## resX.min.nexp 3.279461e-07 3.279461e-07
## resY.min.nexp 6.321031e-13 6.321031e-13
## nImgs.nexp 5.347986e-33 5.347986e-33
## resXY.max.log1p 3.609783e-33 3.609783e-33
## resY.mean.log1p 2.968636e-33 2.968636e-33
## resXY.mean.log1p 2.004916e-33 2.004916e-33
## resX.mean.log1p 1.735618e-33 1.735618e-33
## resX.min.log1p 9.203578e-34 9.203578e-34
## resY.min.log1p 2.901106e-34 2.901106e-34
## resXY.min.log1p 1.803222e-34 1.803222e-34
## resY.mad.log1p 1.172100e-34 1.172100e-34
## resY.mad.nexp 1.093280e-34 1.093280e-34
## resXY.mad.nexp 1.040560e-34 1.040560e-34
## resX.mad.nexp 8.684464e-35 8.684464e-35
## resX.min.root2 7.925067e-35 7.925067e-35
## nImgs.log1p 5.990386e-35 5.990386e-35
## resY.mad.root2 5.370220e-35 5.370220e-35
## nImgs.cut.fctr(120,3e+03] 3.495438e-35 3.495438e-35
## resX.mad.log1p 3.381740e-35 3.381740e-35
## resY.min.root2 3.315149e-35 3.315149e-35
## nImgs.cut.fctr(32,60] 1.322805e-35 1.322805e-35
## resXY.mad.log1p 1.167405e-35 1.167405e-35
## resX.mean.root2 1.054674e-35 1.054674e-35
## nImgs.root2 1.043463e-35 1.043463e-35
## resY.mean.root2 1.040887e-35 1.040887e-35
## resY.mean 8.023804e-36 8.023804e-36
## .rnorm 6.837161e-36 6.837161e-36
## nImgs.cut.fctr(60,120] 6.588626e-36 6.588626e-36
## resXY.max.root2 3.416462e-36 3.416462e-36
## resX.mean 2.937711e-36 2.937711e-36
## resX.mad.root2 1.866320e-36 1.866320e-36
## resX.min 1.122832e-36 1.122832e-36
## resY.mad 1.075883e-36 1.075883e-36
## resXY.min.root2 8.831835e-37 8.831835e-37
## resX.mad 7.534941e-37 7.534941e-37
## resXY.mean.root2 6.317202e-37 6.317202e-37
## resXY.mad.root2 2.943704e-37 2.943704e-37
## resY.min 9.864150e-38 9.864150e-38
## nImgs 1.509626e-38 1.509626e-38
## resXY.mean 1.499542e-38 1.499542e-38
## resXY.max 1.251014e-38 1.251014e-38
## .pos 1.020651e-38 1.020651e-38
## resXY.min 3.259045e-39 3.259045e-39
## resXY.mad 0.000000e+00 0.000000e+00#mdl_id <-"RFE.X.glmnet"; glb_featsimp_df <- myget_feats_importance(glb_models_lst[[mdl_id]], glb_featsimp_df); glb_featsimp_df[, paste0(mdl_id, ".imp")] <- glb_featsimp_df$imp; print(glb_featsimp_df)
#print(head(sbst_featsimp_df <- subset(glb_featsimp_df, is.na(RFE.X.glmnet.imp) | (abs(RFE.X.YeoJohnson.glmnet.imp - RFE.X.glmnet.imp) > 0.0001), select=-imp)))
#print(orderBy(~ -cor.y.abs, subset(glb_feats_df, id %in% c(row.names(sbst_featsimp_df), "startprice.dcm1.is9", "D.weight.post.stop.sum"))))
# Used again in fit.data.training & predict.data.new chunks
glb_analytics_diag_plots <- function(obs_df, mdl_id, prob_threshold=NULL) {
if (!is.null(featsimp_df <- glb_featsimp_df)) {
featsimp_df$feat <- gsub("`(.*?)`", "\\1", row.names(featsimp_df))
featsimp_df$feat.interact <- gsub("(.*?):(.*)", "\\2", featsimp_df$feat)
featsimp_df$feat <- gsub("(.*?):(.*)", "\\1", featsimp_df$feat)
featsimp_df$feat.interact <-
ifelse(featsimp_df$feat.interact == featsimp_df$feat,
NA, featsimp_df$feat.interact)
featsimp_df$feat <-
gsub("(.*?)\\.fctr(.*)", "\\1\\.fctr", featsimp_df$feat)
featsimp_df$feat.interact <-
gsub("(.*?)\\.fctr(.*)", "\\1\\.fctr", featsimp_df$feat.interact)
featsimp_df <- orderBy(~ -imp.max,
summaryBy(imp ~ feat + feat.interact, data=featsimp_df,
FUN=max))
#rex_str=":(.*)"; txt_vctr=tail(featsimp_df$feat); ret_lst <- regexec(rex_str, txt_vctr); ret_lst <- regmatches(txt_vctr, ret_lst); ret_vctr <- sapply(1:length(ret_lst), function(pos_ix) ifelse(length(ret_lst[[pos_ix]]) > 0, ret_lst[[pos_ix]], "")); print(ret_vctr <- ret_vctr[ret_vctr != ""])
featsimp_df <- subset(featsimp_df, !is.na(imp.max))
if (nrow(featsimp_df) > 5) {
warning("Limiting important feature scatter plots to 5 out of ",
nrow(featsimp_df))
featsimp_df <- head(featsimp_df, 5)
}
# if (!all(is.na(featsimp_df$feat.interact)))
# stop("not implemented yet")
rsp_var_out <- mygetPredictIds(glb_rsp_var, mdl_id)$value
for (var in featsimp_df$feat) {
plot_df <- melt(obs_df, id.vars = var,
measure.vars = c(glb_rsp_var, rsp_var_out))
print(myplot_scatter(plot_df, var, "value", colorcol_name = "variable",
facet_colcol_name = "variable", jitter = TRUE) +
guides(color = FALSE))
}
}
if (glb_is_regression) {
if (is.null(featsimp_df) || (nrow(featsimp_df) == 0))
warning("No important features in glb_fin_mdl") else
print(myplot_prediction_regression(df=obs_df,
feat_x=ifelse(nrow(featsimp_df) > 1, featsimp_df$feat[2],
".rownames"),
feat_y=featsimp_df$feat[1],
rsp_var=glb_rsp_var, rsp_var_out=rsp_var_out,
id_vars=glbFeatsId)
# + facet_wrap(reformulate(featsimp_df$feat[2])) # if [1 or 2] is a factor
# + geom_point(aes_string(color="<col_name>.fctr")) # to color the plot
)
}
if (glb_is_classification) {
if (is.null(featsimp_df) || (nrow(featsimp_df) == 0))
warning("No features in selected model are statistically important")
else print(myplot_prediction_classification(df = obs_df,
feat_x = ifelse(nrow(featsimp_df) > 1,
featsimp_df$feat[2], ".rownames"),
feat_y = featsimp_df$feat[1],
rsp_var = glb_rsp_var,
rsp_var_out = rsp_var_out,
id_vars = glbFeatsId,
prob_threshold = prob_threshold))
}
}
if (glb_is_classification && glb_is_binomial)
glb_analytics_diag_plots(obs_df = glbObsOOB, mdl_id = glb_sel_mdl_id,
prob_threshold = glb_models_df[glb_models_df$id == glb_sel_mdl_id,
"opt.prob.threshold.OOB"]) else
glb_analytics_diag_plots(obs_df = glbObsOOB, mdl_id = glb_sel_mdl_id) ## Warning in glb_analytics_diag_plots(obs_df = glbObsOOB, mdl_id =
## glb_sel_mdl_id, : Limiting important feature scatter plots to 5 out of 44
## [1] "Min/Max Boundaries: "
## business_id outdoor.fctr outdoor.fctr.All.X..rcv.glmnet.prob
## 1 710 Y 0.47295347
## 2 286 N 0.07153311
## 3 2846 N 0.99830797
## outdoor.fctr.All.X..rcv.glmnet outdoor.fctr.All.X..rcv.glmnet.err
## 1 Y FALSE
## 2 Y TRUE
## 3 Y TRUE
## outdoor.fctr.All.X..rcv.glmnet.err.abs
## 1 0.52704653
## 2 0.07153311
## 3 0.99830797
## outdoor.fctr.All.X..rcv.glmnet.is.acc
## 1 TRUE
## 2 FALSE
## 3 FALSE
## outdoor.fctr.All.X..rcv.glmnet.accurate
## 1 TRUE
## 2 FALSE
## 3 FALSE
## outdoor.fctr.All.X..rcv.glmnet.error .label
## 1 0.00000000 710
## 2 0.07153311 286
## 3 0.99830797 2846
## [1] "Inaccurate: "
## business_id outdoor.fctr outdoor.fctr.All.X..rcv.glmnet.prob
## 1 3945 N 0.00000000
## 2 1634 N 0.00000000
## 3 286 N 0.07153311
## 4 1402 N 0.15396397
## 5 2495 N 0.17075236
## 6 77 N 0.18236736
## outdoor.fctr.All.X..rcv.glmnet outdoor.fctr.All.X..rcv.glmnet.err
## 1 Y TRUE
## 2 Y TRUE
## 3 Y TRUE
## 4 Y TRUE
## 5 Y TRUE
## 6 Y TRUE
## outdoor.fctr.All.X..rcv.glmnet.err.abs
## 1 0.00000000
## 2 0.00000000
## 3 0.07153311
## 4 0.15396397
## 5 0.17075236
## 6 0.18236736
## outdoor.fctr.All.X..rcv.glmnet.is.acc
## 1 FALSE
## 2 FALSE
## 3 FALSE
## 4 FALSE
## 5 FALSE
## 6 FALSE
## outdoor.fctr.All.X..rcv.glmnet.accurate
## 1 FALSE
## 2 FALSE
## 3 FALSE
## 4 FALSE
## 5 FALSE
## 6 FALSE
## outdoor.fctr.All.X..rcv.glmnet.error
## 1 0.00000000
## 2 0.00000000
## 3 0.07153311
## 4 0.15396397
## 5 0.17075236
## 6 0.18236736
## business_id outdoor.fctr outdoor.fctr.All.X..rcv.glmnet.prob
## 107 2671 N 0.4131585
## 122 3313 N 0.4214526
## 213 3896 N 0.4886172
## 242 3376 N 0.5051562
## 272 2104 N 0.5180997
## 413 3487 N 0.5989197
## outdoor.fctr.All.X..rcv.glmnet outdoor.fctr.All.X..rcv.glmnet.err
## 107 Y TRUE
## 122 Y TRUE
## 213 Y TRUE
## 242 Y TRUE
## 272 Y TRUE
## 413 Y TRUE
## outdoor.fctr.All.X..rcv.glmnet.err.abs
## 107 0.4131585
## 122 0.4214526
## 213 0.4886172
## 242 0.5051562
## 272 0.5180997
## 413 0.5989197
## outdoor.fctr.All.X..rcv.glmnet.is.acc
## 107 FALSE
## 122 FALSE
## 213 FALSE
## 242 FALSE
## 272 FALSE
## 413 FALSE
## outdoor.fctr.All.X..rcv.glmnet.accurate
## 107 FALSE
## 122 FALSE
## 213 FALSE
## 242 FALSE
## 272 FALSE
## 413 FALSE
## outdoor.fctr.All.X..rcv.glmnet.error
## 107 0.4131585
## 122 0.4214526
## 213 0.4886172
## 242 0.5051562
## 272 0.5180997
## 413 0.5989197
## business_id outdoor.fctr outdoor.fctr.All.X..rcv.glmnet.prob
## 492 3980 N 0.7334074
## 493 2646 N 0.7447127
## 494 3018 N 0.7467896
## 495 574 N 0.7489144
## 496 393 N 0.7580598
## 497 2846 N 0.9983080
## outdoor.fctr.All.X..rcv.glmnet outdoor.fctr.All.X..rcv.glmnet.err
## 492 Y TRUE
## 493 Y TRUE
## 494 Y TRUE
## 495 Y TRUE
## 496 Y TRUE
## 497 Y TRUE
## outdoor.fctr.All.X..rcv.glmnet.err.abs
## 492 0.7334074
## 493 0.7447127
## 494 0.7467896
## 495 0.7489144
## 496 0.7580598
## 497 0.9983080
## outdoor.fctr.All.X..rcv.glmnet.is.acc
## 492 FALSE
## 493 FALSE
## 494 FALSE
## 495 FALSE
## 496 FALSE
## 497 FALSE
## outdoor.fctr.All.X..rcv.glmnet.accurate
## 492 FALSE
## 493 FALSE
## 494 FALSE
## 495 FALSE
## 496 FALSE
## 497 FALSE
## outdoor.fctr.All.X..rcv.glmnet.error
## 492 0.7334074
## 493 0.7447127
## 494 0.7467896
## 495 0.7489144
## 496 0.7580598
## 497 0.9983080
if (!is.null(glbFeatsCategory)) {
glbLvlCategory <- merge(glbLvlCategory,
myget_category_stats(obs_df = glbObsFit, mdl_id = glb_sel_mdl_id,
label = "fit"),
by = glbFeatsCategory, all = TRUE)
row.names(glbLvlCategory) <- glbLvlCategory[, glbFeatsCategory]
glbLvlCategory <- merge(glbLvlCategory,
myget_category_stats(obs_df = glbObsOOB, mdl_id = glb_sel_mdl_id,
label="OOB"),
#by=glbFeatsCategory, all=TRUE) glb_ctgry-df already contains .n.OOB ?
all = TRUE)
row.names(glbLvlCategory) <- glbLvlCategory[, glbFeatsCategory]
if (any(grepl("OOB", glbMdlMetricsEval)))
print(orderBy(~-err.abs.OOB.mean, glbLvlCategory)) else
print(orderBy(~-err.abs.fit.mean, glbLvlCategory))
print(colSums(glbLvlCategory[, -grep(glbFeatsCategory, names(glbLvlCategory))]))
}## nImgs.cut.fctr .n.OOB .n.Fit .n.Tst .freqRatio.Fit
## (32,60] (32,60] 243 278 2512 0.2774451
## (60,120] (60,120] 260 257 2459 0.2564870
## (0,32] (0,32] 237 238 2532 0.2375250
## (120,3e+03] (120,3e+03] 258 229 2497 0.2285429
## .freqRatio.OOB .freqRatio.Tst err.abs.fit.sum err.abs.fit.mean
## (32,60] 0.2434870 0.2512 134.8544 0.4850879
## (60,120] 0.2605210 0.2459 123.8999 0.4821008
## (0,32] 0.2374749 0.2532 109.1638 0.4586716
## (120,3e+03] 0.2585170 0.2497 108.0508 0.4718376
## .n.fit err.abs.OOB.sum err.abs.OOB.mean
## (32,60] 278 122.3399 0.5034566
## (60,120] 257 129.2920 0.4972768
## (0,32] 238 117.2849 0.4948730
## (120,3e+03] 229 126.8451 0.4916477
## .n.OOB .n.Fit .n.Tst .freqRatio.Fit
## 998.000000 1002.000000 10000.000000 1.000000
## .freqRatio.OOB .freqRatio.Tst err.abs.fit.sum err.abs.fit.mean
## 1.000000 1.000000 475.969008 1.897698
## .n.fit err.abs.OOB.sum err.abs.OOB.mean
## 1002.000000 495.761912 1.987254write.csv(glbObsOOB[, c(glbFeatsId,
grep(glb_rsp_var, names(glbObsOOB), fixed=TRUE, value=TRUE))],
paste0(gsub(".", "_", paste0(glbOut$pfx, glb_sel_mdl_id), fixed=TRUE),
"_OOBobs.csv"), row.names=FALSE)
fit.models_2_chunk_df <-
myadd_chunk(NULL, "fit.models_2_bgn", label.minor = "teardown")## label step_major step_minor label_minor bgn end elapsed
## 1 fit.models_2_bgn 1 0 teardown 224.745 NA NAglb_chunks_df <- myadd_chunk(glb_chunks_df, "fit.models", major.inc=FALSE)## label step_major step_minor label_minor bgn end elapsed
## 18 fit.models 8 2 2 213.842 224.756 10.914
## 19 fit.models 8 3 3 224.757 NA NA# if (sum(is.na(glbObsAll$D.P.http)) > 0)
# stop("fit.models_3: Why is this happening ?")
#stop(here"); glb2Sav()
sync_glb_obs_df <- function() {
# Merge or cbind ?
for (col in setdiff(names(glbObsFit), names(glbObsTrn)))
glbObsTrn[glbObsTrn$.lcn == "Fit", col] <<- glbObsFit[, col]
for (col in setdiff(names(glbObsFit), names(glbObsAll)))
glbObsAll[glbObsAll$.lcn == "Fit", col] <<- glbObsFit[, col]
if (all(is.na(glbObsNew[, glb_rsp_var])))
for (col in setdiff(names(glbObsOOB), names(glbObsTrn)))
glbObsTrn[glbObsTrn$.lcn == "OOB", col] <<- glbObsOOB[, col]
for (col in setdiff(names(glbObsOOB), names(glbObsAll)))
glbObsAll[glbObsAll$.lcn == "OOB", col] <<- glbObsOOB[, col]
}
sync_glb_obs_df()
print(setdiff(names(glbObsNew), names(glbObsAll)))## character(0)replay.petrisim(pn=glb_analytics_pn,
replay.trans=(glb_analytics_avl_objs <- c(glb_analytics_avl_objs,
"model.selected")), flip_coord=TRUE)## time trans "bgn " "fit.data.training.all " "predict.data.new " "end "
## 0.0000 multiple enabled transitions: data.training.all data.new model.selected firing: data.training.all
## 1.0000 1 2 1 0 0
## 1.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction firing: data.new
## 2.0000 2 1 1 1 0
## 2.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction data.new.prediction firing: model.selected
## 3.0000 3 0 2 1 0
glb_chunks_df <- myadd_chunk(glb_chunks_df, "fit.data.training", major.inc = TRUE)## label step_major step_minor label_minor bgn end
## 19 fit.models 8 3 3 224.757 229.013
## 20 fit.data.training 9 0 0 229.013 NA
## elapsed
## 19 4.256
## 20 NAStep 9.0: fit data training
#load(paste0(glb_inp_pfx, "dsk.RData"))
if (!is.null(glb_fin_mdl_id) && (glb_fin_mdl_id %in% names(glb_models_lst))) {
warning("Final model same as user selected model")
glb_fin_mdl <- glb_models_lst[[glb_fin_mdl_id]]
} else
# if (nrow(glbObsFit) + length(glbObsFitOutliers) == nrow(glbObsTrn))
if (!all(is.na(glbObsNew[, glb_rsp_var])))
{
warning("Final model same as glb_sel_mdl_id")
glb_fin_mdl_id <- paste0("Final.", glb_sel_mdl_id)
glb_fin_mdl <- glb_sel_mdl
glb_models_lst[[glb_fin_mdl_id]] <- glb_fin_mdl
} else {
if (grepl("RFE\\.X", names(glbMdlFamilies))) {
indep_vars <- myadjust_interaction_feats(subset(glb_feats_df,
!nzv & (exclude.as.feat != 1))[, "id"])
rfe_trn_results <-
myrun_rfe(glbObsTrn, indep_vars, glbRFESizes[["Final"]])
if (!isTRUE(all.equal(sort(predictors(rfe_trn_results)),
sort(predictors(rfe_fit_results))))) {
print("Diffs predictors(rfe_trn_results) vs. predictors(rfe_fit_results):")
print(setdiff(predictors(rfe_trn_results), predictors(rfe_fit_results)))
print("Diffs predictors(rfe_fit_results) vs. predictors(rfe_trn_results):")
print(setdiff(predictors(rfe_fit_results), predictors(rfe_trn_results)))
}
}
# }
if (grepl("Ensemble", glb_sel_mdl_id)) {
# Find which models are relevant
mdlimp_df <- subset(myget_feats_importance(glb_sel_mdl), imp > 5)
# Fit selected models on glbObsTrn
for (mdl_id in gsub(".prob", "",
gsub(mygetPredictIds(glb_rsp_var)$value, "", row.names(mdlimp_df), fixed = TRUE),
fixed = TRUE)) {
mdl_id_components <- unlist(strsplit(mdl_id, "[.]"))
mdlIdPfx <- paste0(c(head(mdl_id_components, -1), "Train"),
collapse = ".")
if (grepl("RFE\\.X\\.", mdlIdPfx))
mdlIndepVars <- myadjust_interaction_feats(myextract_actual_feats(
predictors(rfe_trn_results))) else
mdlIndepVars <- trim(unlist(
strsplit(glb_models_df[glb_models_df$id == mdl_id, "feats"], "[,]")))
ret_lst <-
myfit_mdl(mdl_specs_lst = myinit_mdl_specs_lst(mdl_specs_lst = list(
id.prefix = mdlIdPfx,
type = glb_model_type, tune.df = glbMdlTuneParams,
trainControl.method = "repeatedcv",
trainControl.number = glb_rcv_n_folds,
trainControl.repeats = glb_rcv_n_repeats,
trainControl.classProbs = glb_is_classification,
trainControl.summaryFunction = glbMdlMetricSummaryFn,
train.metric = glbMdlMetricSummary,
train.maximize = glbMdlMetricMaximize,
train.method = tail(mdl_id_components, 1))),
indep_vars = mdlIndepVars,
rsp_var = glb_rsp_var,
fit_df = glbObsTrn, OOB_df = NULL)
glbObsTrn <- glb_get_predictions(df = glbObsTrn,
mdl_id = tail(glb_models_df$id, 1),
rsp_var = glb_rsp_var,
prob_threshold_def =
subset(glb_models_df, id == mdl_id)$opt.prob.threshold.OOB)
glbObsNew <- glb_get_predictions(df = glbObsNew,
mdl_id = tail(glb_models_df$id, 1),
rsp_var = glb_rsp_var,
prob_threshold_def =
subset(glb_models_df, id == mdl_id)$opt.prob.threshold.OOB)
}
}
# "Final" model
if ((model_method <- glb_sel_mdl$method) == "custom")
# get actual method from the mdl_id
model_method <- tail(unlist(strsplit(glb_sel_mdl_id, "[.]")), 1)
if (grepl("Ensemble", glb_sel_mdl_id)) {
# Find which models are relevant
mdlimp_df <- subset(myget_feats_importance(glb_sel_mdl), imp > 5)
if (glb_is_classification && glb_is_binomial)
indep_vars_vctr <- gsub("(.*)\\.(.*)\\.prob", "\\1\\.Train\\.\\2\\.prob",
row.names(mdlimp_df)) else
indep_vars_vctr <- gsub("(.*)\\.(.*)", "\\1\\.Train\\.\\2",
row.names(mdlimp_df))
} else
if (grepl("RFE.X", glb_sel_mdl_id, fixed = TRUE)) {
indep_vars_vctr <- myextract_actual_feats(predictors(rfe_trn_results))
} else indep_vars_vctr <-
trim(unlist(strsplit(glb_models_df[glb_models_df$id ==
glb_sel_mdl_id
, "feats"], "[,]")))
if (!is.null(glb_preproc_methods) &&
((match_pos <- regexpr(gsub(".", "\\.",
paste(glb_preproc_methods, collapse = "|"),
fixed = TRUE), glb_sel_mdl_id)) != -1))
ths_preProcess <- str_sub(glb_sel_mdl_id, match_pos,
match_pos + attr(match_pos, "match.length") - 1) else
ths_preProcess <- NULL
mdl_id_pfx <- ifelse(grepl("Ensemble", glb_sel_mdl_id),
"Final.Ensemble", "Final")
trnobs_df <- glbObsTrn
if (!is.null(glbObsTrnOutliers[[mdl_id_pfx]])) {
trnobs_df <- glbObsTrn[!(glbObsTrn[, glbFeatsId] %in% glbObsTrnOutliers[[mdl_id_pfx]]), ]
print(sprintf("Outliers removed: %d", nrow(glbObsTrn) - nrow(trnobs_df)))
print(setdiff(glbObsTrn[, glbFeatsId], trnobs_df[, glbFeatsId]))
}
# Force fitting of Final.glm to identify outliers
method_vctr <- unique(c(myparseMdlId(glb_sel_mdl_id)$alg, glbMdlFamilies[["Final"]]))
for (method in method_vctr) {
#source("caret_nominalTrainWorkflow.R")
# glmnet requires at least 2 indep vars
if ((length(indep_vars_vctr) == 1) && (method %in% "glmnet"))
next
ret_lst <-
myfit_mdl(mdl_specs_lst = myinit_mdl_specs_lst(mdl_specs_lst = list(
id.prefix = mdl_id_pfx,
type = glb_model_type, trainControl.method = "repeatedcv",
trainControl.number = glb_rcv_n_folds,
trainControl.repeats = glb_rcv_n_repeats,
trainControl.classProbs = glb_is_classification,
trainControl.summaryFunction = glbMdlMetricSummaryFn,
trainControl.allowParallel = glbMdlAllowParallel,
train.metric = glbMdlMetricSummary,
train.maximize = glbMdlMetricMaximize,
train.method = method,
train.preProcess = ths_preProcess)),
indep_vars = indep_vars_vctr, rsp_var = glb_rsp_var,
fit_df = trnobs_df, OOB_df = NULL)
if ((length(method_vctr) == 1) || (method != "glm")) {
glb_fin_mdl <- glb_models_lst[[length(glb_models_lst)]]
glb_fin_mdl_id <- glb_models_df[length(glb_models_lst), "id"]
}
}
}## [1] "myfit_mdl: enter: 0.000000 secs"
## [1] "fitting model: Final##rcv#glmnet"
## [1] " indep_vars: nImgs.cut.fctr,nImgs.log1p,.pos,resX.mad.log1p,resX.mad.root2,resX.mad,resXY.mad.nexp,nImgs.root2,resY.mean.log1p,resY.mean.root2,resY.mean,resY.mad.nexp,resY.mad,resXY.max.log1p,resXY.max.root2,resXY.max,resY.mad.root2,resY.mad.log1p,nImgs.nexp,resXY.mean.log1p,resXY.mean.root2,.rnorm,resXY.mean,resXY.mad.root2,resXY.mad,resX.mad.nexp,resXY.mad.log1p,nImgs,resX.mean.log1p,resX.mean.root2,resX.mean,resX.min.nexp,resX.mean.nexp,resY.mean.nexp,resY.min.nexp,resX.min.log1p,resX.min.root2,resX.min,resXY.min.log1p,resXY.min.root2,resY.min.log1p,resY.min.root2,resXY.min,resY.min"
## [1] "myfit_mdl: setup complete: 0.738000 secs"
## Aggregating results
## Selecting tuning parameters
## Fitting alpha = 1, lambda = 0.00236 on full training set
## [1] "myfit_mdl: train complete: 22.713000 secs"## Warning in myfit_mdl(mdl_specs_lst = myinit_mdl_specs_lst(mdl_specs_lst
## = list(id.prefix = mdl_id_pfx, : model's bestTune found at an extreme of
## tuneGrid for parameter: alpha
## Length Class Mode
## a0 50 -none- numeric
## beta 2300 dgCMatrix S4
## df 50 -none- numeric
## dim 2 -none- numeric
## lambda 50 -none- numeric
## dev.ratio 50 -none- numeric
## nulldev 1 -none- numeric
## npasses 1 -none- numeric
## jerr 1 -none- numeric
## offset 1 -none- logical
## classnames 2 -none- character
## call 5 -none- call
## nobs 1 -none- numeric
## lambdaOpt 1 -none- numeric
## xNames 46 -none- character
## problemType 1 -none- character
## tuneValue 2 data.frame list
## obsLevels 2 -none- character
## [1] "min lambda > lambdaOpt:"
## (Intercept) .pos .rnorm
## -3.063087e+00 7.035722e-05 -5.454361e-03
## nImgs nImgs.cut.fctr(32,60] nImgs.log1p
## -9.785444e-04 1.150942e-01 2.553484e-01
## nImgs.nexp resX.mad.log1p resX.mean.nexp
## 1.920825e+00 9.324753e-03 -9.900000e+35
## resX.min resX.min.nexp resXY.max
## 2.707822e-04 -6.916709e+26 -2.975932e-06
## resXY.mean resXY.min resY.mad
## -1.801540e-07 -1.103577e-06 1.487094e-04
## resY.mad.nexp resY.mean.log1p resY.mean.nexp
## 2.248354e-02 4.798175e-01 -9.900000e+35
## resY.min resY.min.nexp
## -5.930960e-04 -9.302222e+12
## [1] "max lambda < lambdaOpt:"
## (Intercept) .pos .rnorm
## -3.344275e+00 7.183591e-05 -6.355870e-03
## nImgs nImgs.cut.fctr(32,60] nImgs.log1p
## -1.002766e-03 1.181961e-01 2.620960e-01
## nImgs.nexp resX.mad.log1p resX.mean.nexp
## 2.259889e+00 9.110044e-03 -9.900000e+35
## resX.min resX.min.nexp resXY.max
## 3.549463e-04 -7.206478e+26 -3.073427e-06
## resXY.mean resXY.min resY.mad
## -4.916887e-07 -1.216240e-06 2.113887e-04
## resY.mad.nexp resY.mean.log1p resY.mean.nexp
## 2.948711e-02 5.314093e-01 -9.900000e+35
## resY.min resY.min.nexp
## -5.776209e-04 -9.728555e+12
## [1] "myfit_mdl: train diagnostics complete: 23.305000 secs"
## Prediction
## Reference N Y
## N 42 955
## Y 19 984
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 5.130000e-01 2.324849e-02 4.908389e-01 5.351230e-01 5.015000e-01
## AccuracyPValue McnemarPValue
## 1.571511e-01 3.325830e-197
## [1] "myfit_mdl: predict complete: 25.347000 secs"
## id
## 1 Final##rcv#glmnet
## feats
## 1 nImgs.cut.fctr,nImgs.log1p,.pos,resX.mad.log1p,resX.mad.root2,resX.mad,resXY.mad.nexp,nImgs.root2,resY.mean.log1p,resY.mean.root2,resY.mean,resY.mad.nexp,resY.mad,resXY.max.log1p,resXY.max.root2,resXY.max,resY.mad.root2,resY.mad.log1p,nImgs.nexp,resXY.mean.log1p,resXY.mean.root2,.rnorm,resXY.mean,resXY.mad.root2,resXY.mad,resX.mad.nexp,resXY.mad.log1p,nImgs,resX.mean.log1p,resX.mean.root2,resX.mean,resX.min.nexp,resX.mean.nexp,resY.mean.nexp,resY.min.nexp,resX.min.log1p,resX.min.root2,resX.min,resXY.min.log1p,resXY.min.root2,resY.min.log1p,resY.min.root2,resXY.min,resY.min
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 25 21.952 0.35
## max.AUCpROC.fit max.Sens.fit max.Spec.fit max.AUCROCR.fit
## 1 0.5367168 0.442327 0.6311067 0.559857
## opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.4 0.6689327 0.5109885
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit
## 1 0.4908389 0.535123 0.02161427
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.01801986 0.03607316
## [1] "myfit_mdl: exit: 25.363000 secs"rm(ret_lst)
glb_chunks_df <- myadd_chunk(glb_chunks_df, "fit.data.training", major.inc=FALSE)## label step_major step_minor label_minor bgn end
## 20 fit.data.training 9 0 0 229.013 254.864
## 21 fit.data.training 9 1 1 254.865 NA
## elapsed
## 20 25.851
## 21 NA#stop(here"); glb2Sav()
if (glb_is_classification && glb_is_binomial)
prob_threshold <- glb_models_df[glb_models_df$id == glb_sel_mdl_id,
"opt.prob.threshold.OOB"] else
prob_threshold <- NULL
if (grepl("Ensemble", glb_fin_mdl_id)) {
# Get predictions for each model in ensemble; Outliers that have been moved to OOB might not have been predicted yet
mdlEnsembleComps <- unlist(str_split(subset(glb_models_df,
id == glb_fin_mdl_id)$feats, ","))
if (glb_is_classification && glb_is_binomial)
mdlEnsembleComps <- gsub("\\.prob$", "", mdlEnsembleComps)
mdlEnsembleComps <- gsub(paste0("^",
gsub(".", "\\.", mygetPredictIds(glb_rsp_var)$value, fixed = TRUE)),
"", mdlEnsembleComps)
for (mdl_id in mdlEnsembleComps) {
glbObsTrn <- glb_get_predictions(df = glbObsTrn, mdl_id = mdl_id,
rsp_var = glb_rsp_var,
prob_threshold_def = prob_threshold)
glbObsNew <- glb_get_predictions(df = glbObsNew, mdl_id = mdl_id,
rsp_var = glb_rsp_var,
prob_threshold_def = prob_threshold)
}
}
glbObsTrn <- glb_get_predictions(df = glbObsTrn, mdl_id = glb_fin_mdl_id,
rsp_var = glb_rsp_var,
prob_threshold_def = prob_threshold)## Warning in glb_get_predictions(df = glbObsTrn, mdl_id = glb_fin_mdl_id, :
## Using default probability threshold: 0glb_featsimp_df <- myget_feats_importance(mdl=glb_fin_mdl,
featsimp_df=glb_featsimp_df)
#glb_featsimp_df[, paste0(glb_fin_mdl_id, ".imp")] <- glb_featsimp_df$imp
print(glb_featsimp_df)## All.X..rcv.glmnet.imp Final..rcv.glmnet.imp
## resX.mean.nexp 1.000000e+02 1.000000e+02
## resY.mean.nexp 1.000000e+02 1.000000e+02
## resX.min.nexp 3.279461e-07 7.150779e-08
## resY.min.nexp 6.321031e-13 9.637774e-22
## nImgs.nexp 5.347986e-33 2.132364e-34
## resY.mean.log1p 2.968636e-33 5.138997e-35
## nImgs.log1p 5.990386e-35 2.617513e-35
## nImgs.cut.fctr(32,60] 1.322805e-35 1.180146e-35
## resY.mad.nexp 1.093280e-34 2.667936e-36
## resX.mad.log1p 3.381740e-35 9.297274e-37
## .rnorm 6.837161e-36 6.020313e-37
## nImgs 1.509626e-38 1.002154e-37
## resY.min 9.864150e-38 5.903176e-38
## resX.min 1.122832e-36 3.212106e-38
## resY.mad 1.075883e-36 1.857300e-38
## .pos 1.020651e-38 7.190583e-39
## resXY.max 1.251014e-38 3.061240e-40
## resXY.min 3.259045e-39 1.178567e-40
## resXY.mean 1.499542e-38 3.585110e-41
## nImgs.cut.fctr(120,3e+03] 3.495438e-35 0.000000e+00
## nImgs.cut.fctr(60,120] 6.588626e-36 0.000000e+00
## nImgs.root2 1.043463e-35 0.000000e+00
## resX.mad 7.534941e-37 0.000000e+00
## resX.mad.nexp 8.684464e-35 0.000000e+00
## resX.mad.root2 1.866320e-36 0.000000e+00
## resX.mean 2.937711e-36 0.000000e+00
## resX.mean.log1p 1.735618e-33 0.000000e+00
## resX.mean.root2 1.054674e-35 0.000000e+00
## resX.min.log1p 9.203578e-34 0.000000e+00
## resX.min.root2 7.925067e-35 0.000000e+00
## resXY.mad 0.000000e+00 0.000000e+00
## resXY.mad.log1p 1.167405e-35 0.000000e+00
## resXY.mad.nexp 1.040560e-34 0.000000e+00
## resXY.mad.root2 2.943704e-37 0.000000e+00
## resXY.max.log1p 3.609783e-33 0.000000e+00
## resXY.max.root2 3.416462e-36 0.000000e+00
## resXY.mean.log1p 2.004916e-33 0.000000e+00
## resXY.mean.root2 6.317202e-37 0.000000e+00
## resXY.min.log1p 1.803222e-34 0.000000e+00
## resXY.min.root2 8.831835e-37 0.000000e+00
## resY.mad.log1p 1.172100e-34 0.000000e+00
## resY.mad.root2 5.370220e-35 0.000000e+00
## resY.mean 8.023804e-36 0.000000e+00
## resY.mean.root2 1.040887e-35 0.000000e+00
## resY.min.log1p 2.901106e-34 0.000000e+00
## resY.min.root2 3.315149e-35 0.000000e+00
## imp
## resX.mean.nexp 1.000000e+02
## resY.mean.nexp 1.000000e+02
## resX.min.nexp 7.150779e-08
## resY.min.nexp 9.637774e-22
## nImgs.nexp 2.132364e-34
## resY.mean.log1p 5.138997e-35
## nImgs.log1p 2.617513e-35
## nImgs.cut.fctr(32,60] 1.180146e-35
## resY.mad.nexp 2.667936e-36
## resX.mad.log1p 9.297274e-37
## .rnorm 6.020313e-37
## nImgs 1.002154e-37
## resY.min 5.903176e-38
## resX.min 3.212106e-38
## resY.mad 1.857300e-38
## .pos 7.190583e-39
## resXY.max 3.061240e-40
## resXY.min 1.178567e-40
## resXY.mean 3.585110e-41
## nImgs.cut.fctr(120,3e+03] 0.000000e+00
## nImgs.cut.fctr(60,120] 0.000000e+00
## nImgs.root2 0.000000e+00
## resX.mad 0.000000e+00
## resX.mad.nexp 0.000000e+00
## resX.mad.root2 0.000000e+00
## resX.mean 0.000000e+00
## resX.mean.log1p 0.000000e+00
## resX.mean.root2 0.000000e+00
## resX.min.log1p 0.000000e+00
## resX.min.root2 0.000000e+00
## resXY.mad 0.000000e+00
## resXY.mad.log1p 0.000000e+00
## resXY.mad.nexp 0.000000e+00
## resXY.mad.root2 0.000000e+00
## resXY.max.log1p 0.000000e+00
## resXY.max.root2 0.000000e+00
## resXY.mean.log1p 0.000000e+00
## resXY.mean.root2 0.000000e+00
## resXY.min.log1p 0.000000e+00
## resXY.min.root2 0.000000e+00
## resY.mad.log1p 0.000000e+00
## resY.mad.root2 0.000000e+00
## resY.mean 0.000000e+00
## resY.mean.root2 0.000000e+00
## resY.min.log1p 0.000000e+00
## resY.min.root2 0.000000e+00if (glb_is_classification && glb_is_binomial)
glb_analytics_diag_plots(obs_df=glbObsTrn, mdl_id=glb_fin_mdl_id,
prob_threshold=glb_models_df[glb_models_df$id == glb_sel_mdl_id,
"opt.prob.threshold.OOB"]) else
glb_analytics_diag_plots(obs_df=glbObsTrn, mdl_id=glb_fin_mdl_id) ## Warning in glb_analytics_diag_plots(obs_df = glbObsTrn, mdl_id =
## glb_fin_mdl_id, : Limiting important feature scatter plots to 5 out of 44
## [1] "Min/Max Boundaries: "
## business_id outdoor.fctr outdoor.fctr.All.X..rcv.glmnet.prob
## 1 710 Y NA
## 2 2846 N NA
## 3 286 N NA
## outdoor.fctr.All.X..rcv.glmnet outdoor.fctr.All.X..rcv.glmnet.err
## 1 <NA> NA
## 2 <NA> NA
## 3 <NA> NA
## outdoor.fctr.All.X..rcv.glmnet.err.abs
## 1 NA
## 2 NA
## 3 NA
## outdoor.fctr.All.X..rcv.glmnet.is.acc
## 1 NA
## 2 NA
## 3 NA
## outdoor.fctr.Final..rcv.glmnet.prob outdoor.fctr.Final..rcv.glmnet
## 1 0.4019945 Y
## 2 0.3788056 Y
## 3 0.3900328 Y
## outdoor.fctr.Final..rcv.glmnet.err
## 1 FALSE
## 2 TRUE
## 3 TRUE
## outdoor.fctr.Final..rcv.glmnet.err.abs
## 1 0.5980055
## 2 0.3788056
## 3 0.3900328
## outdoor.fctr.Final..rcv.glmnet.is.acc
## 1 TRUE
## 2 FALSE
## 3 FALSE
## outdoor.fctr.Final..rcv.glmnet.accurate
## 1 TRUE
## 2 FALSE
## 3 FALSE
## outdoor.fctr.Final..rcv.glmnet.error .label
## 1 0.0000000 710
## 2 0.3788056 2846
## 3 0.3900328 286
## [1] "Inaccurate: "
## business_id outdoor.fctr outdoor.fctr.All.X..rcv.glmnet.prob
## 1 1634 N NA
## 2 1445 N 0.0007709958
## 3 2339 N 0.4216553578
## 4 1114 N 0.2697178532
## 5 3065 N 0.1225230650
## 6 2611 N 0.2918475560
## outdoor.fctr.All.X..rcv.glmnet outdoor.fctr.All.X..rcv.glmnet.err
## 1 <NA> NA
## 2 Y TRUE
## 3 Y TRUE
## 4 Y TRUE
## 5 Y TRUE
## 6 Y TRUE
## outdoor.fctr.All.X..rcv.glmnet.err.abs
## 1 NA
## 2 0.0007709958
## 3 0.4216553578
## 4 0.2697178532
## 5 0.1225230650
## 6 0.2918475560
## outdoor.fctr.All.X..rcv.glmnet.is.acc
## 1 NA
## 2 FALSE
## 3 FALSE
## 4 FALSE
## 5 FALSE
## 6 FALSE
## outdoor.fctr.Final..rcv.glmnet.prob outdoor.fctr.Final..rcv.glmnet
## 1 0.1055817 Y
## 2 0.1065867 Y
## 3 0.1372385 Y
## 4 0.1392711 Y
## 5 0.2485974 Y
## 6 0.2693462 Y
## outdoor.fctr.Final..rcv.glmnet.err
## 1 TRUE
## 2 TRUE
## 3 TRUE
## 4 TRUE
## 5 TRUE
## 6 TRUE
## outdoor.fctr.Final..rcv.glmnet.err.abs
## 1 0.1055817
## 2 0.1065867
## 3 0.1372385
## 4 0.1392711
## 5 0.2485974
## 6 0.2693462
## outdoor.fctr.Final..rcv.glmnet.is.acc
## 1 FALSE
## 2 FALSE
## 3 FALSE
## 4 FALSE
## 5 FALSE
## 6 FALSE
## outdoor.fctr.Final..rcv.glmnet.accurate
## 1 FALSE
## 2 FALSE
## 3 FALSE
## 4 FALSE
## 5 FALSE
## 6 FALSE
## outdoor.fctr.Final..rcv.glmnet.error
## 1 0.1055817
## 2 0.1065867
## 3 0.1372385
## 4 0.1392711
## 5 0.2485974
## 6 0.2693462
## business_id outdoor.fctr outdoor.fctr.All.X..rcv.glmnet.prob
## 97 3849 N NA
## 231 2971 N NA
## 266 1796 N NA
## 267 3403 N NA
## 440 590 N 0.5656691
## 732 1378 N NA
## outdoor.fctr.All.X..rcv.glmnet outdoor.fctr.All.X..rcv.glmnet.err
## 97 <NA> NA
## 231 <NA> NA
## 266 <NA> NA
## 267 <NA> NA
## 440 Y TRUE
## 732 <NA> NA
## outdoor.fctr.All.X..rcv.glmnet.err.abs
## 97 NA
## 231 NA
## 266 NA
## 267 NA
## 440 0.5656691
## 732 NA
## outdoor.fctr.All.X..rcv.glmnet.is.acc
## 97 NA
## 231 NA
## 266 NA
## 267 NA
## 440 FALSE
## 732 NA
## outdoor.fctr.Final..rcv.glmnet.prob outdoor.fctr.Final..rcv.glmnet
## 97 0.4253303 Y
## 231 0.4673646 Y
## 266 0.4744031 Y
## 267 0.4744499 Y
## 440 0.4998805 Y
## 732 0.5297045 Y
## outdoor.fctr.Final..rcv.glmnet.err
## 97 TRUE
## 231 TRUE
## 266 TRUE
## 267 TRUE
## 440 TRUE
## 732 TRUE
## outdoor.fctr.Final..rcv.glmnet.err.abs
## 97 0.4253303
## 231 0.4673646
## 266 0.4744031
## 267 0.4744499
## 440 0.4998805
## 732 0.5297045
## outdoor.fctr.Final..rcv.glmnet.is.acc
## 97 FALSE
## 231 FALSE
## 266 FALSE
## 267 FALSE
## 440 FALSE
## 732 FALSE
## outdoor.fctr.Final..rcv.glmnet.accurate
## 97 FALSE
## 231 FALSE
## 266 FALSE
## 267 FALSE
## 440 FALSE
## 732 FALSE
## outdoor.fctr.Final..rcv.glmnet.error
## 97 0.4253303
## 231 0.4673646
## 266 0.4744031
## 267 0.4744499
## 440 0.4998805
## 732 0.5297045
## business_id outdoor.fctr outdoor.fctr.All.X..rcv.glmnet.prob
## 992 615 N 0.4612589
## 993 955 N NA
## 994 3737 N 0.3893913
## 995 941 N NA
## 996 3858 N 0.2536333
## 997 3285 N 0.7940934
## outdoor.fctr.All.X..rcv.glmnet outdoor.fctr.All.X..rcv.glmnet.err
## 992 Y TRUE
## 993 <NA> NA
## 994 Y TRUE
## 995 <NA> NA
## 996 Y TRUE
## 997 Y TRUE
## outdoor.fctr.All.X..rcv.glmnet.err.abs
## 992 0.4612589
## 993 NA
## 994 0.3893913
## 995 NA
## 996 0.2536333
## 997 0.7940934
## outdoor.fctr.All.X..rcv.glmnet.is.acc
## 992 FALSE
## 993 NA
## 994 FALSE
## 995 NA
## 996 FALSE
## 997 FALSE
## outdoor.fctr.Final..rcv.glmnet.prob outdoor.fctr.Final..rcv.glmnet
## 992 0.5890333 Y
## 993 0.5920903 Y
## 994 0.5930546 Y
## 995 0.5957571 Y
## 996 0.6010328 Y
## 997 0.6023388 Y
## outdoor.fctr.Final..rcv.glmnet.err
## 992 TRUE
## 993 TRUE
## 994 TRUE
## 995 TRUE
## 996 TRUE
## 997 TRUE
## outdoor.fctr.Final..rcv.glmnet.err.abs
## 992 0.5890333
## 993 0.5920903
## 994 0.5930546
## 995 0.5957571
## 996 0.6010328
## 997 0.6023388
## outdoor.fctr.Final..rcv.glmnet.is.acc
## 992 FALSE
## 993 FALSE
## 994 FALSE
## 995 FALSE
## 996 FALSE
## 997 FALSE
## outdoor.fctr.Final..rcv.glmnet.accurate
## 992 FALSE
## 993 FALSE
## 994 FALSE
## 995 FALSE
## 996 FALSE
## 997 FALSE
## outdoor.fctr.Final..rcv.glmnet.error
## 992 0.5890333
## 993 0.5920903
## 994 0.5930546
## 995 0.5957571
## 996 0.6010328
## 997 0.6023388
dsp_feats_vctr <- c(NULL)
for(var in grep(".imp", names(glb_feats_df), fixed=TRUE, value=TRUE))
dsp_feats_vctr <- union(dsp_feats_vctr,
glb_feats_df[!is.na(glb_feats_df[, var]), "id"])
# print(glbObsTrn[glbObsTrn$UniqueID %in% FN_OOB_ids,
# grep(glb_rsp_var, names(glbObsTrn), value=TRUE)])
print(setdiff(names(glbObsTrn), names(glbObsAll)))## [1] "outdoor.fctr.Final..rcv.glmnet.prob"
## [2] "outdoor.fctr.Final..rcv.glmnet"
## [3] "outdoor.fctr.Final..rcv.glmnet.err"
## [4] "outdoor.fctr.Final..rcv.glmnet.err.abs"
## [5] "outdoor.fctr.Final..rcv.glmnet.is.acc"for (col in setdiff(names(glbObsTrn), names(glbObsAll)))
# Merge or cbind ?
glbObsAll[glbObsAll$.src == "Train", col] <- glbObsTrn[, col]
print(setdiff(names(glbObsFit), names(glbObsAll)))## character(0)print(setdiff(names(glbObsOOB), names(glbObsAll)))## character(0)for (col in setdiff(names(glbObsOOB), names(glbObsAll)))
# Merge or cbind ?
glbObsAll[glbObsAll$.lcn == "OOB", col] <- glbObsOOB[, col]
print(setdiff(names(glbObsNew), names(glbObsAll)))## character(0)#glb2Sav(); all.equal(savObsAll, glbObsAll); all.equal(sav_models_lst, glb_models_lst)
#load(file = paste0(glbOut$pfx, "dsk_knitr.RData"))
#cmpCols <- names(glbObsAll)[!grepl("\\.Final\\.", names(glbObsAll))]; all.equal(savObsAll[, cmpCols], glbObsAll[, cmpCols]); all.equal(savObsAll[, "H.P.http"], glbObsAll[, "H.P.http"]);
replay.petrisim(pn = glb_analytics_pn,
replay.trans = (glb_analytics_avl_objs <- c(glb_analytics_avl_objs,
"data.training.all.prediction","model.final")), flip_coord = TRUE)## time trans "bgn " "fit.data.training.all " "predict.data.new " "end "
## 0.0000 multiple enabled transitions: data.training.all data.new model.selected firing: data.training.all
## 1.0000 1 2 1 0 0
## 1.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction firing: data.new
## 2.0000 2 1 1 1 0
## 2.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction data.new.prediction firing: model.selected
## 3.0000 3 0 2 1 0
## 3.0000 multiple enabled transitions: model.final data.training.all.prediction data.new.prediction firing: data.training.all.prediction
## 4.0000 5 0 1 1 1
## 4.0000 multiple enabled transitions: model.final data.training.all.prediction data.new.prediction firing: model.final
## 5.0000 4 0 0 2 1
glb_chunks_df <- myadd_chunk(glb_chunks_df, "predict.data.new", major.inc = TRUE)## label step_major step_minor label_minor bgn end
## 21 fit.data.training 9 1 1 254.865 261.304
## 22 predict.data.new 10 0 0 261.304 NA
## elapsed
## 21 6.439
## 22 NAStep 10.0: predict data new
## Warning in glb_get_predictions(obs_df, mdl_id = glb_fin_mdl_id, rsp_var =
## glb_rsp_var, : Using default probability threshold: 0
## Warning in glb_get_predictions(obs_df, mdl_id = glb_fin_mdl_id, rsp_var =
## glb_rsp_var, : Using default probability threshold: 0## Warning in glb_analytics_diag_plots(obs_df = glbObsNew, mdl_id =
## glb_fin_mdl_id, : Limiting important feature scatter plots to 5 out of 44## Warning: Removed 10000 rows containing missing values (geom_point).
## Warning: Removed 10000 rows containing missing values (geom_point).
## Warning: Removed 10000 rows containing missing values (geom_point).
## Warning: Removed 10000 rows containing missing values (geom_point).
## Warning: Removed 10000 rows containing missing values (geom_point).
## Warning: Removed 10000 rows containing missing values (geom_point).
## Warning: Removed 10000 rows containing missing values (geom_point).
## Warning: Removed 10000 rows containing missing values (geom_point).
## Warning: Removed 10000 rows containing missing values (geom_point).
## Warning: Removed 10000 rows containing missing values (geom_point).
## NULL## Loading required package: tidyr##
## Attaching package: 'tidyr'## The following object is masked from 'package:Matrix':
##
## expand## [1] "OOBobs outdoor.fctr.All.X..rcv.glmnet Y: min < min of Train range: 25"
## business_id outdoor.fctr.All.X..rcv.glmnet .pos resX.mean
## 2 1001 Y 2 449.2222
## 254 1402 Y 254 329.3333
## 662 2146 Y 662 414.7059
## 879 2562 Y 879 492.8571
## 1027 2846 Y 1027 500.0000
## 1036 286 Y 1036 283.7692
## resX.mean.log1p resX.mean.nexp resX.mean.root2 resX.min.nexp
## 2 6.109741 8.040216e-196 21.19486 3.331006e-156
## 254 5.800102 9.383193e-144 18.14754 3.342796e-80
## 662 6.029978 7.861831e-181 20.36433 5.148200e-131
## 879 6.202246 9.012857e-215 22.20039 1.915170e-174
## 1027 6.216606 7.124576e-218 22.36068 7.124576e-218
## 1036 5.651679 5.762208e-124 16.84545 2.053885e-85
## resXY.mean resXY.mean.log1p resXY.mean.root2 resXY.min
## 2 187964.00 12.14401 433.5482 179000
## 254 115110.00 11.65365 339.2786 27985
## 662 144558.82 11.88145 380.2089 67500
## 879 168107.14 12.03236 410.0087 120000
## 1027 191250.00 12.16134 437.3214 187500
## 1036 87626.15 11.38085 296.0172 25350
## resXY.min.log1p resXY.min.root2 resY.mad.nexp resY.mean
## 2 12.09515 423.0839 9.674088e-10 426.4444
## 254 10.23946 167.2872 3.857670e-36 322.5333
## 662 11.11990 259.8076 4.477805e-85 335.5882
## 879 11.69526 346.4102 3.008950e-14 340.5000
## 1027 12.14154 433.0127 1.482049e-05 382.5000
## 1036 10.14057 159.2168 1.000000e+00 253.0000
## resY.mean.log1p resY.mean.nexp resY.mean.root2 resY.min
## 2 6.057824 6.273804e-186 20.65053 361
## 254 5.779302 8.424674e-141 17.95921 145
## 662 5.818860 1.802525e-146 18.31907 225
## 879 5.833348 1.326566e-148 18.45264 281
## 1027 5.949340 7.627122e-167 19.55761 375
## 1036 5.537334 1.328912e-110 15.90597 129
## resY.min.log1p resY.min.nexp resY.min.root2
## 2 5.891644 1.658410e-157 19.00000
## 254 4.983607 1.064879e-63 12.04159
## 662 5.420535 1.921948e-98 15.00000
## 879 5.641907 9.188626e-123 16.76305
## 1027 5.929589 1.379016e-163 19.36492
## 1036 4.867534 9.462629e-57 11.35782
## business_id outdoor.fctr.All.X..rcv.glmnet .pos resX.mean
## 254 1402 Y 254 329.3333
## 1027 2846 Y 1027 500.0000
## 14 1026 Y 14 428.2785
## 1383 3521 Y 1383 409.6552
## 1630 3946 Y 1630 419.5556
## 6 101 Y 6 433.4380
## resX.mean.log1p resX.mean.nexp resX.mean.root2 resX.min.nexp
## 254 5.800102 9.383193e-144 18.14754 3.342796e-80
## 1027 6.216606 7.124576e-218 22.36068 7.124576e-218
## 14 6.062106 1.002349e-186 20.69489 9.188626e-123
## 1383 6.017754 1.227493e-178 20.23994 2.937482e-30
## 1630 6.041577 6.156549e-183 20.48306 4.817492e-144
## 6 6.074053 5.757861e-189 20.81917 1.399426e-130
## resXY.mean resXY.mean.log1p resXY.mean.root2 resXY.min
## 254 115110.0 11.65365 339.2786 27985
## 1027 191250.0 12.16134 437.3214 187500
## 14 180094.9 12.10124 424.3759 131000
## 1383 180536.4 12.10369 424.8958 4556
## 1630 202740.7 12.21969 450.2674 165000
## 6 184607.1 12.12599 429.6593 124848
## resXY.min.log1p resXY.min.root2 resY.mad.nexp resY.mean
## 254 10.23946 167.28718 3.857670e-36 322.5333
## 1027 12.14154 433.01270 1.482049e-05 382.5000
## 14 11.78296 361.93922 1.000000e+00 431.9114
## 1383 8.42442 67.49815 1.000000e+00 438.5172
## 1630 12.01371 406.20192 1.000000e+00 485.9259
## 6 11.73486 353.33836 1.000000e+00 435.3471
## resY.mean.log1p resY.mean.nexp resY.mean.root2 resY.min
## 254 5.779302 8.424674e-141 17.95921 145
## 1027 5.949340 7.627122e-167 19.55761 375
## 14 6.070533 2.650120e-188 20.78248 262
## 1383 6.085677 3.584114e-191 20.94080 67
## 1630 6.188112 9.226814e-212 22.04373 370
## 6 6.078438 8.534017e-190 20.86497 282
## resY.min.log1p resY.min.nexp resY.min.root2
## 254 4.983607 1.064879e-63 12.041595
## 1027 5.929589 1.379016e-163 19.364917
## 14 5.572154 1.640007e-114 16.186414
## 1383 4.219508 7.984904e-30 8.185353
## 1630 5.916202 2.046641e-161 19.235384
## 6 5.645447 3.380307e-123 16.792856
## business_id outdoor.fctr.All.X..rcv.glmnet .pos resX.mean
## 7 1011 Y 7 412.1000
## 13 1024 Y 13 462.6364
## 586 2007 Y 586 401.5652
## 6 101 Y 6 433.4380
## 12 1022 Y 12 453.3919
## 748 2315 Y 748 437.1244
## resX.mean.log1p resX.mean.nexp resX.mean.root2 resX.min.nexp
## 7 6.023690 1.064741e-179 20.30025 2.162867e-105
## 13 6.139101 1.201103e-201 21.50898 6.519766e-145
## 586 5.997857 4.003520e-175 20.03909 1.404379e-54
## 6 6.074053 5.757861e-189 20.81917 1.399426e-130
## 12 6.118960 1.242824e-197 21.29300 9.188626e-123
## 748 6.082503 1.442998e-190 20.90752 2.678637e-33
## resXY.mean resXY.mean.log1p resXY.mean.root2 resXY.min resXY.min.log1p
## 7 179669.2 12.09888 423.8740 87001 11.37369
## 13 183931.8 12.12233 428.8727 166000 12.01975
## 586 180321.6 12.10250 424.6429 7065 8.86305
## 6 184607.1 12.12599 429.6593 124848 11.73486
## 12 184123.6 12.12337 429.0963 76800 11.24897
## 748 180895.3 12.10568 425.3179 3675 8.20958
## resXY.min.root2 resY.mad.nexp resY.mean resY.mean.log1p resY.mean.nexp
## 7 294.95932 1.000000e+00 445.6286 6.101727 2.923953e-194
## 13 407.43098 9.053782e-28 405.2273 6.006913 1.028091e-176
## 586 84.05355 1.000000e+00 447.9565 6.106926 2.850719e-195
## 6 353.33836 1.000000e+00 435.3471 6.078438 8.534017e-190
## 12 277.12813 1.160754e-49 414.1229 6.028575 1.408288e-180
## 748 60.62178 6.775247e-62 419.7319 6.041996 5.161464e-183
## resY.mean.root2 resY.min resY.min.log1p resY.min.nexp resY.min.root2
## 7 21.10992 274 5.616771 1.007655e-119 16.552945
## 13 20.13026 332 5.808142 6.519766e-145 18.220867
## 586 21.16498 45 3.828641 2.862519e-20 6.708204
## 6 20.86497 282 5.645447 3.380307e-123 16.792856
## 12 20.35001 240 5.484797 5.879283e-105 15.491933
## 748 20.48736 49 3.912023 5.242886e-22 7.000000
## id cor.y exclude.as.feat cor.y.abs
## .pos .pos 0.027497300 FALSE 0.027497300
## resX.mean resX.mean -0.017726551 FALSE 0.017726551
## resX.mean.log1p resX.mean.log1p -0.015059015 FALSE 0.015059015
## resX.mean.nexp resX.mean.nexp -0.022433472 FALSE 0.022433472
## resX.mean.root2 resX.mean.root2 -0.016434019 FALSE 0.016434019
## resX.min.nexp resX.min.nexp -0.022391602 FALSE 0.022391602
## resXY.mean resXY.mean -0.009002880 FALSE 0.009002880
## resXY.mean.log1p resXY.mean.log1p -0.004867571 FALSE 0.004867571
## resXY.mean.root2 resXY.mean.root2 -0.007039955 FALSE 0.007039955
## resXY.min resXY.min -0.049458217 FALSE 0.049458217
## resXY.min.log1p resXY.min.log1p -0.033756424 FALSE 0.033756424
## resXY.min.root2 resXY.min.root2 -0.041449898 FALSE 0.041449898
## resY.mad.nexp resY.mad.nexp 0.012190340 FALSE 0.012190340
## resY.mean resY.mean 0.012599188 FALSE 0.012599188
## resY.mean.log1p resY.mean.log1p 0.013625190 FALSE 0.013625190
## resY.mean.nexp resY.mean.nexp -0.022433472 FALSE 0.022433472
## resY.mean.root2 resY.mean.root2 0.013106506 FALSE 0.013106506
## resY.min resY.min -0.050925308 FALSE 0.050925308
## resY.min.log1p resY.min.log1p -0.043072548 FALSE 0.043072548
## resY.min.nexp resY.min.nexp -0.022433600 FALSE 0.022433600
## resY.min.root2 resY.min.root2 -0.047387777 FALSE 0.047387777
## cor.high.X freqRatio percentUnique zeroVar nzv
## .pos <NA> 1.000000 100.00 FALSE FALSE
## resX.mean <NA> 2.000000 97.75 FALSE FALSE
## resX.mean.log1p resX.mean 2.000000 97.60 FALSE FALSE
## resX.mean.nexp <NA> 2.000000 97.75 FALSE FALSE
## resX.mean.root2 resX.mean 2.000000 97.45 FALSE FALSE
## resX.min.nexp <NA> 6.000000 11.45 FALSE FALSE
## resXY.mean <NA> 6.000000 98.55 FALSE FALSE
## resXY.mean.log1p <NA> 4.000000 90.80 FALSE FALSE
## resXY.mean.root2 <NA> 6.000000 98.20 FALSE FALSE
## resXY.min resY.min 9.745455 37.65 FALSE FALSE
## resXY.min.log1p resXY.min 9.745455 37.65 FALSE FALSE
## resXY.min.root2 resXY.min 9.745455 37.65 FALSE FALSE
## resY.mad.nexp <NA> 5.354497 9.05 FALSE FALSE
## resY.mean resY.mean.root2 1.666667 98.15 FALSE FALSE
## resY.mean.log1p <NA> 1.666667 97.90 FALSE FALSE
## resY.mean.nexp resX.mean.nexp 1.666667 98.15 FALSE FALSE
## resY.mean.root2 resY.mean.log1p 1.666667 97.85 FALSE FALSE
## resY.min <NA> 9.824561 13.85 FALSE FALSE
## resY.min.log1p resY.min 9.824561 13.85 FALSE FALSE
## resY.min.nexp <NA> 9.824561 13.85 FALSE FALSE
## resY.min.root2 resY.min 9.824561 13.85 FALSE FALSE
## is.cor.y.abs.low interaction.feat shapiro.test.p.value
## .pos FALSE NA 2.145811e-24
## resX.mean FALSE NA 1.161337e-19
## resX.mean.log1p FALSE NA 2.973500e-25
## resX.mean.nexp FALSE NA 1.194234e-72
## resX.mean.root2 FALSE NA 1.959497e-22
## resX.min.nexp FALSE NA 1.195403e-72
## resXY.mean FALSE NA 2.964553e-36
## resXY.mean.log1p TRUE NA 6.980019e-43
## resXY.mean.root2 TRUE NA 1.780045e-39
## resXY.min FALSE NA 2.084930e-32
## resXY.min.log1p FALSE NA 1.076069e-42
## resXY.min.root2 FALSE NA 1.752753e-35
## resY.mad.nexp FALSE NA 1.839563e-53
## resY.mean FALSE NA 1.464051e-21
## resY.mean.log1p FALSE NA 3.854130e-28
## resY.mean.nexp FALSE NA 1.194234e-72
## resY.mean.root2 FALSE NA 7.216658e-25
## resY.min FALSE NA 1.973528e-28
## resY.min.log1p FALSE NA 3.088017e-38
## resY.min.nexp FALSE NA 1.194238e-72
## resY.min.root2 FALSE NA 2.137435e-32
## rsp_var_raw id_var rsp_var max min
## .pos FALSE NA NA 1.200000e+04 1.000000e+00
## resX.mean FALSE NA NA 5.000000e+02 2.837692e+02
## resX.mean.log1p FALSE NA NA 6.216606e+00 5.651679e+00
## resX.mean.nexp FALSE NA NA 5.762208e-124 7.124576e-218
## resX.mean.root2 FALSE NA NA 2.236068e+01 1.684545e+01
## resX.min.nexp FALSE NA NA 9.602680e-24 7.124576e-218
## resXY.mean FALSE NA NA 2.500000e+05 8.762615e+04
## resXY.mean.log1p FALSE NA NA 1.242922e+01 1.138085e+01
## resXY.mean.root2 FALSE NA NA 5.000000e+02 2.960172e+02
## resXY.min FALSE NA NA 2.500000e+05 1.802000e+03
## resXY.min.log1p FALSE NA NA 1.242922e+01 7.497207e+00
## resXY.min.root2 FALSE NA NA 5.000000e+02 4.244997e+01
## resY.mad.nexp FALSE NA NA 1.000000e+00 8.904719e-122
## resY.mean FALSE NA NA 5.000000e+02 2.530000e+02
## resY.mean.log1p FALSE NA NA 6.216606e+00 5.537334e+00
## resY.mean.nexp FALSE NA NA 1.328912e-110 7.124576e-218
## resY.mean.root2 FALSE NA NA 2.236068e+01 1.590597e+01
## resY.min FALSE NA NA 5.000000e+02 2.900000e+01
## resY.min.log1p FALSE NA NA 6.216606e+00 3.401197e+00
## resY.min.nexp FALSE NA NA 2.543666e-13 7.124576e-218
## resY.min.root2 FALSE NA NA 2.236068e+01 5.385165e+00
## max.outdoor.fctr.N max.outdoor.fctr.Y min.outdoor.fctr.N
## .pos 2.000000e+03 1.996000e+03 3.000000e+00
## resX.mean 4.910714e+02 4.940000e+02 3.553137e+02
## resX.mean.log1p 6.198624e+00 6.204558e+00 5.875812e+00
## resX.mean.nexp 4.888883e-155 1.209672e-151 5.375122e-214
## resX.mean.root2 2.216013e+01 2.222611e+01 1.884977e+01
## resX.min.nexp 3.221340e-27 2.937482e-30 1.379016e-163
## resXY.mean 2.087027e+05 2.011129e+05 1.348537e+05
## resXY.mean.log1p 1.224867e+01 1.221163e+01 1.181195e+01
## resXY.mean.root2 4.568399e+02 4.484562e+02 3.672243e+02
## resXY.min 1.875000e+05 1.875000e+05 5.000000e+03
## resXY.min.log1p 1.214154e+01 1.214154e+01 8.517393e+00
## resXY.min.root2 4.330127e+02 4.330127e+02 7.071068e+01
## resY.mad.nexp 1.000000e+00 1.000000e+00 3.268701e-81
## resY.mean 4.847273e+02 4.851765e+02 3.266000e+02
## resY.mean.log1p 6.185647e+00 6.186572e+00 5.791793e+00
## resY.mean.nexp 1.443518e-142 2.530221e-127 3.059287e-211
## resY.mean.root2 2.201652e+01 2.202672e+01 1.807208e+01
## resY.min 3.750000e+02 3.750000e+02 5.000000e+01
## resY.min.log1p 5.929589e+00 5.929589e+00 3.931826e+00
## resY.min.nexp 1.928750e-22 7.095474e-23 1.379016e-163
## resY.min.root2 1.936492e+01 1.936492e+01 7.071068e+00
## min.outdoor.fctr.Y max.outdoor.fctr.All.X..rcv.glmnet.Y
## .pos 1.500000e+01 1.999000e+03
## resX.mean 3.475000e+02 5.000000e+02
## resX.mean.log1p 5.853638e+00 6.216606e+00
## resX.mean.nexp 2.874259e-215 5.762208e-124
## resX.mean.root2 1.864135e+01 2.236068e+01
## resX.min.nexp 1.379016e-163 2.937482e-30
## resXY.mean 1.137250e+05 2.182778e+05
## resXY.mean.log1p 1.164155e+01 1.229353e+01
## resXY.mean.root2 3.372314e+02 4.672021e+02
## resXY.min 4.624000e+03 1.875000e+05
## resXY.min.log1p 8.439232e+00 1.214154e+01
## resXY.min.root2 6.800000e+01 4.330127e+02
## resY.mad.nexp 3.268701e-81 1.000000e+00
## resY.mean 2.915000e+02 5.000000e+02
## resY.mean.log1p 5.678465e+00 6.216606e+00
## resY.mean.nexp 1.952253e-211 1.328912e-110
## resY.mean.root2 1.707337e+01 2.236068e+01
## resY.min 5.100000e+01 5.000000e+02
## resY.min.log1p 3.951244e+00 6.216606e+00
## resY.min.nexp 1.379016e-163 2.543666e-13
## resY.min.root2 7.141428e+00 2.236068e+01
## min.outdoor.fctr.All.X..rcv.glmnet.Y
## .pos 1.000000e+00
## resX.mean 2.837692e+02
## resX.mean.log1p 5.651679e+00
## resX.mean.nexp 7.124576e-218
## resX.mean.root2 1.684545e+01
## resX.min.nexp 7.124576e-218
## resXY.mean 8.762615e+04
## resXY.mean.log1p 1.138085e+01
## resXY.mean.root2 2.960172e+02
## resXY.min 3.675000e+03
## resXY.min.log1p 8.209580e+00
## resXY.min.root2 6.062178e+01
## resY.mad.nexp 4.477805e-85
## resY.mean 2.530000e+02
## resY.mean.log1p 5.537334e+00
## resY.mean.nexp 7.124576e-218
## resY.mean.root2 1.590597e+01
## resY.min 2.900000e+01
## resY.min.log1p 3.401197e+00
## resY.min.nexp 7.124576e-218
## resY.min.root2 5.385165e+00
## max.outdoor.fctr.Final..rcv.glmnet.Y
## .pos 1.200000e+04
## resX.mean 5.000000e+02
## resX.mean.log1p 6.216606e+00
## resX.mean.nexp 5.719134e-133
## resX.mean.root2 2.236068e+01
## resX.min.nexp 9.602680e-24
## resXY.mean 2.500000e+05
## resXY.mean.log1p 1.242922e+01
## resXY.mean.root2 5.000000e+02
## resXY.min 2.500000e+05
## resXY.min.log1p 1.242922e+01
## resXY.min.root2 5.000000e+02
## resY.mad.nexp 1.000000e+00
## resY.mean 5.000000e+02
## resY.mean.log1p 6.216606e+00
## resY.mean.nexp 1.105028e-116
## resY.mean.root2 2.236068e+01
## resY.min 5.000000e+02
## resY.min.log1p 6.216606e+00
## resY.min.nexp 2.543666e-13
## resY.min.root2 2.236068e+01
## min.outdoor.fctr.Final..rcv.glmnet.Y
## .pos 2.001000e+03
## resX.mean 3.045000e+02
## resX.mean.log1p 5.721950e+00
## resX.mean.nexp 7.124576e-218
## resX.mean.root2 1.744993e+01
## resX.min.nexp 7.124576e-218
## resXY.mean 1.058460e+05
## resXY.mean.log1p 1.156975e+01
## resXY.mean.root2 3.253398e+02
## resXY.min 1.802000e+03
## resXY.min.log1p 7.497207e+00
## resXY.min.root2 4.244997e+01
## resY.mad.nexp 8.904719e-122
## resY.mean 2.670000e+02
## resY.mean.log1p 5.590987e+00
## resY.mean.nexp 7.124576e-218
## resY.mean.root2 1.634013e+01
## resY.min 2.900000e+01
## resY.min.log1p 3.401197e+00
## resY.min.nexp 7.124576e-218
## resY.min.root2 5.385165e+00
## [1] "OOBobs outdoor.fctr.All.X..rcv.glmnet Y: max > max of Train range: 28"
## business_id outdoor.fctr.All.X..rcv.glmnet .pos resX.mean
## 254 1402 Y 254 329.3333
## 662 2146 Y 662 414.7059
## 879 2562 Y 879 492.8571
## 962 2719 Y 962 446.1111
## 1027 2846 Y 1027 500.0000
## 1036 286 Y 1036 283.7692
## resX.mean.log1p resX.mean.nexp resX.mean.root2 resX.min
## 254 5.800102 9.383193e-144 18.14754 183
## 662 6.029978 7.861831e-181 20.36433 300
## 879 6.202246 9.012857e-215 22.20039 400
## 962 6.102807 1.804705e-194 21.12134 282
## 1027 6.216606 7.124576e-218 22.36068 500
## 1036 5.651679 5.762208e-124 16.84545 195
## resX.min.log1p resX.min.root2 resXY.mean resXY.mean.log1p
## 254 5.214936 13.52775 115110.00 11.65365
## 662 5.707110 17.32051 144558.82 11.88145
## 879 5.993961 20.00000 168107.14 12.03236
## 962 5.645447 16.79286 201222.22 12.21217
## 1027 6.216606 22.36068 191250.00 12.16134
## 1036 5.278115 13.96424 87626.15 11.38085
## resXY.mean.root2 resY.mad resY.mad.log1p resY.mad.root2 resY.mean
## 254 339.2786 81.5430 4.413319 9.030116 322.5333
## 662 380.2089 194.2206 5.274130 13.936305 335.5882
## 879 410.0087 31.1346 3.469933 5.579839 340.5000
## 962 448.5780 0.0000 0.000000 0.000000 456.3333
## 1027 437.3214 11.1195 2.494816 3.334591 382.5000
## 1036 296.0172 0.0000 0.000000 0.000000 253.0000
## resY.mean.log1p resY.mean.nexp resY.mean.root2 resY.min
## 254 5.779302 8.424674e-141 17.95921 145
## 662 5.818860 1.802525e-146 18.31907 225
## 879 5.833348 1.326566e-148 18.45264 281
## 962 6.125413 6.560719e-199 21.36196 373
## 1027 5.949340 7.627122e-167 19.55761 375
## 1036 5.537334 1.328912e-110 15.90597 129
## resY.min.log1p resY.min.nexp resY.min.root2
## 254 4.983607 1.064879e-63 12.04159
## 662 5.420535 1.921948e-98 15.00000
## 879 5.641907 9.188626e-123 16.76305
## 962 5.924256 1.018963e-162 19.31321
## 1027 5.929589 1.379016e-163 19.36492
## 1036 4.867534 9.462629e-57 11.35782
## business_id outdoor.fctr.All.X..rcv.glmnet .pos resX.mean
## 879 2562 Y 879 492.8571
## 875 2557 Y 875 453.4444
## 849 2495 Y 849 472.4044
## 1999 998 Y 1999 442.0125
## 43 1069 Y 43 441.0667
## 1165 310 Y 1165 500.0000
## resX.mean.log1p resX.mean.nexp resX.mean.root2 resX.min
## 879 6.202246 9.012857e-215 22.20039 400
## 875 6.119076 1.179180e-197 21.29424 281
## 849 6.159950 6.876552e-206 21.73487 281
## 1999 6.093598 1.087453e-192 21.02409 281
## 43 6.091461 2.800145e-192 21.00159 373
## 1165 6.216606 7.124576e-218 22.36068 500
## resX.min.log1p resX.min.root2 resXY.mean resXY.mean.log1p
## 879 5.993961 20.00000 168107.1 12.03236
## 875 5.641907 16.76305 202777.8 12.21987
## 849 5.641907 16.76305 217574.0 12.29030
## 1999 5.641907 16.76305 183271.9 12.11873
## 43 5.924256 19.31321 203733.3 12.22457
## 1165 6.216606 22.36068 190333.3 12.15654
## resXY.mean.root2 resY.mad resY.mad.log1p resY.mad.root2 resY.mean
## 879 410.0087 31.1346 3.469933 5.579839 340.5000
## 875 450.3085 0.0000 0.000000 0.000000 452.1111
## 849 466.4483 0.0000 0.000000 0.000000 462.6838
## 1999 428.1026 139.3644 4.944242 11.805270 424.5312
## 43 451.3683 0.0000 0.000000 0.000000 466.4000
## 1165 436.2721 0.0000 0.000000 0.000000 380.6667
## resY.mean.log1p resY.mean.nexp resY.mean.root2 resY.min
## 879 5.833348 1.326566e-148 18.45264 281
## 875 6.116137 4.473415e-197 21.26290 281
## 849 6.139203 1.145430e-201 21.51009 259
## 1999 6.053338 4.250311e-185 20.60416 279
## 43 6.147185 2.786465e-203 21.59630 373
## 1165 5.944548 4.770536e-166 19.51068 373
## resY.min.log1p resY.min.nexp resY.min.root2
## 879 5.641907 9.188626e-123 16.76305
## 875 5.641907 9.188626e-123 16.76305
## 849 5.560682 3.294042e-113 16.09348
## 1999 5.634790 6.789527e-122 16.70329
## 43 5.924256 1.018963e-162 19.31321
## 1165 5.924256 1.018963e-162 19.31321
## business_id outdoor.fctr.All.X..rcv.glmnet .pos resX.mean
## 1054 2895 Y 1054 415.3617
## 1997 993 Y 1997 428.4412
## 553 1944 Y 553 452.2698
## 586 2007 Y 586 401.5652
## 599 2031 Y 599 448.7042
## 748 2315 Y 748 437.1244
## resX.mean.log1p resX.mean.nexp resX.mean.root2 resX.min
## 1054 6.031554 4.080419e-181 20.38042 119
## 1997 6.062485 8.518463e-187 20.69882 208
## 553 6.116488 3.816837e-197 21.26664 281
## 586 5.997857 4.003520e-175 20.03909 124
## 599 6.108590 1.349680e-195 21.18264 281
## 748 6.082503 1.442998e-190 20.90752 75
## resX.min.log1p resX.min.root2 resXY.mean resXY.mean.log1p
## 1054 4.787492 10.908712 189951.8 12.15453
## 1997 5.342334 14.422205 172134.1 12.05604
## 553 5.641907 16.763055 201468.3 12.21339
## 586 4.828314 11.135529 180321.6 12.10250
## 599 5.641907 16.763055 202484.5 12.21842
## 748 4.330733 8.660254 180895.3 12.10568
## resXY.mean.root2 resY.mad resY.mad.log1p resY.mad.root2 resY.mean
## 1054 435.8346 0.0000 0.000000 0.00000 457.1489
## 1997 414.8905 140.1057 4.949509 11.83663 412.7353
## 553 448.8522 0.0000 0.000000 0.00000 450.6667
## 586 424.6429 0.0000 0.000000 0.00000 447.9565
## 599 449.9827 0.0000 0.000000 0.00000 456.1268
## 748 425.3179 140.8470 4.954749 11.86790 419.7319
## resY.mean.log1p resY.mean.nexp resY.mean.root2 resY.min
## 1054 6.127194 2.902282e-199 21.38104 44
## 1997 6.025226 5.640782e-180 20.31589 280
## 553 6.112944 1.896503e-196 21.22891 281
## 586 6.106926 2.850719e-195 21.16498 45
## 599 6.124961 8.066123e-199 21.35712 281
## 748 6.041996 5.161464e-183 20.48736 49
## resY.min.log1p resY.min.nexp resY.min.root2
## 1054 3.806662 7.781132e-20 6.633250
## 1997 5.638355 2.497728e-122 16.733201
## 553 5.641907 9.188626e-123 16.763055
## 586 3.828641 2.862519e-20 6.708204
## 599 5.641907 9.188626e-123 16.763055
## 748 3.912023 5.242886e-22 7.000000
## id cor.y exclude.as.feat cor.y.abs
## .pos .pos 0.027497300 FALSE 0.027497300
## resX.mean resX.mean -0.017726551 FALSE 0.017726551
## resX.mean.log1p resX.mean.log1p -0.015059015 FALSE 0.015059015
## resX.mean.nexp resX.mean.nexp -0.022433472 FALSE 0.022433472
## resX.mean.root2 resX.mean.root2 -0.016434019 FALSE 0.016434019
## resX.min resX.min -0.031436275 FALSE 0.031436275
## resX.min.log1p resX.min.log1p -0.030103276 FALSE 0.030103276
## resX.min.root2 resX.min.root2 -0.030339745 FALSE 0.030339745
## resXY.mean resXY.mean -0.009002880 FALSE 0.009002880
## resXY.mean.log1p resXY.mean.log1p -0.004867571 FALSE 0.004867571
## resXY.mean.root2 resXY.mean.root2 -0.007039955 FALSE 0.007039955
## resY.mad resY.mad 0.007630633 FALSE 0.007630633
## resY.mad.log1p resY.mad.log1p -0.001526058 FALSE 0.001526058
## resY.mad.root2 resY.mad.root2 0.002557583 FALSE 0.002557583
## resY.mean resY.mean 0.012599188 FALSE 0.012599188
## resY.mean.log1p resY.mean.log1p 0.013625190 FALSE 0.013625190
## resY.mean.nexp resY.mean.nexp -0.022433472 FALSE 0.022433472
## resY.mean.root2 resY.mean.root2 0.013106506 FALSE 0.013106506
## resY.min resY.min -0.050925308 FALSE 0.050925308
## resY.min.log1p resY.min.log1p -0.043072548 FALSE 0.043072548
## resY.min.nexp resY.min.nexp -0.022433600 FALSE 0.022433600
## resY.min.root2 resY.min.root2 -0.047387777 FALSE 0.047387777
## cor.high.X freqRatio percentUnique zeroVar nzv
## .pos <NA> 1.000000 100.00 FALSE FALSE
## resX.mean <NA> 2.000000 97.75 FALSE FALSE
## resX.mean.log1p resX.mean 2.000000 97.60 FALSE FALSE
## resX.mean.nexp <NA> 2.000000 97.75 FALSE FALSE
## resX.mean.root2 resX.mean 2.000000 97.45 FALSE FALSE
## resX.min <NA> 6.000000 11.45 FALSE FALSE
## resX.min.log1p resX.min 6.000000 11.45 FALSE FALSE
## resX.min.root2 resX.min 6.000000 11.45 FALSE FALSE
## resXY.mean <NA> 6.000000 98.55 FALSE FALSE
## resXY.mean.log1p <NA> 4.000000 90.80 FALSE FALSE
## resXY.mean.root2 <NA> 6.000000 98.20 FALSE FALSE
## resY.mad <NA> 5.354497 9.05 FALSE FALSE
## resY.mad.log1p <NA> 5.354497 9.05 FALSE FALSE
## resY.mad.root2 <NA> 5.354497 9.05 FALSE FALSE
## resY.mean resY.mean.root2 1.666667 98.15 FALSE FALSE
## resY.mean.log1p <NA> 1.666667 97.90 FALSE FALSE
## resY.mean.nexp resX.mean.nexp 1.666667 98.15 FALSE FALSE
## resY.mean.root2 resY.mean.log1p 1.666667 97.85 FALSE FALSE
## resY.min <NA> 9.824561 13.85 FALSE FALSE
## resY.min.log1p resY.min 9.824561 13.85 FALSE FALSE
## resY.min.nexp <NA> 9.824561 13.85 FALSE FALSE
## resY.min.root2 resY.min 9.824561 13.85 FALSE FALSE
## is.cor.y.abs.low interaction.feat shapiro.test.p.value
## .pos FALSE NA 2.145811e-24
## resX.mean FALSE NA 1.161337e-19
## resX.mean.log1p FALSE NA 2.973500e-25
## resX.mean.nexp FALSE NA 1.194234e-72
## resX.mean.root2 FALSE NA 1.959497e-22
## resX.min FALSE NA 2.978480e-35
## resX.min.log1p FALSE NA 1.543786e-43
## resX.min.root2 FALSE NA 7.580680e-39
## resXY.mean FALSE NA 2.964553e-36
## resXY.mean.log1p TRUE NA 6.980019e-43
## resXY.mean.root2 TRUE NA 1.780045e-39
## resY.mad TRUE NA 3.711302e-48
## resY.mad.log1p TRUE NA 3.133148e-49
## resY.mad.root2 TRUE NA 1.717662e-47
## resY.mean FALSE NA 1.464051e-21
## resY.mean.log1p FALSE NA 3.854130e-28
## resY.mean.nexp FALSE NA 1.194234e-72
## resY.mean.root2 FALSE NA 7.216658e-25
## resY.min FALSE NA 1.973528e-28
## resY.min.log1p FALSE NA 3.088017e-38
## resY.min.nexp FALSE NA 1.194238e-72
## resY.min.root2 FALSE NA 2.137435e-32
## rsp_var_raw id_var rsp_var max min
## .pos FALSE NA NA 1.200000e+04 1.000000e+00
## resX.mean FALSE NA NA 5.000000e+02 2.837692e+02
## resX.mean.log1p FALSE NA NA 6.216606e+00 5.651679e+00
## resX.mean.nexp FALSE NA NA 5.762208e-124 7.124576e-218
## resX.mean.root2 FALSE NA NA 2.236068e+01 1.684545e+01
## resX.min FALSE NA NA 5.000000e+02 5.300000e+01
## resX.min.log1p FALSE NA NA 6.216606e+00 3.988984e+00
## resX.min.root2 FALSE NA NA 2.236068e+01 7.280110e+00
## resXY.mean FALSE NA NA 2.500000e+05 8.762615e+04
## resXY.mean.log1p FALSE NA NA 1.242922e+01 1.138085e+01
## resXY.mean.root2 FALSE NA NA 5.000000e+02 2.960172e+02
## resY.mad FALSE NA NA 2.787288e+02 0.000000e+00
## resY.mad.log1p FALSE NA NA 5.633821e+00 0.000000e+00
## resY.mad.root2 FALSE NA NA 1.669517e+01 0.000000e+00
## resY.mean FALSE NA NA 5.000000e+02 2.530000e+02
## resY.mean.log1p FALSE NA NA 6.216606e+00 5.537334e+00
## resY.mean.nexp FALSE NA NA 1.328912e-110 7.124576e-218
## resY.mean.root2 FALSE NA NA 2.236068e+01 1.590597e+01
## resY.min FALSE NA NA 5.000000e+02 2.900000e+01
## resY.min.log1p FALSE NA NA 6.216606e+00 3.401197e+00
## resY.min.nexp FALSE NA NA 2.543666e-13 7.124576e-218
## resY.min.root2 FALSE NA NA 2.236068e+01 5.385165e+00
## max.outdoor.fctr.N max.outdoor.fctr.Y min.outdoor.fctr.N
## .pos 2.000000e+03 1.996000e+03 3.000000e+00
## resX.mean 4.910714e+02 4.940000e+02 3.553137e+02
## resX.mean.log1p 6.198624e+00 6.204558e+00 5.875812e+00
## resX.mean.nexp 4.888883e-155 1.209672e-151 5.375122e-214
## resX.mean.root2 2.216013e+01 2.222611e+01 1.884977e+01
## resX.min 3.750000e+02 3.750000e+02 6.100000e+01
## resX.min.log1p 5.929589e+00 5.929589e+00 4.127134e+00
## resX.min.root2 1.936492e+01 1.936492e+01 7.810250e+00
## resXY.mean 2.087027e+05 2.011129e+05 1.348537e+05
## resXY.mean.log1p 1.224867e+01 1.221163e+01 1.181195e+01
## resXY.mean.root2 4.568399e+02 4.484562e+02 3.672243e+02
## resY.mad 1.853250e+02 1.853250e+02 0.000000e+00
## resY.mad.log1p 5.227492e+00 5.227492e+00 0.000000e+00
## resY.mad.root2 1.361341e+01 1.361341e+01 0.000000e+00
## resY.mean 4.847273e+02 4.851765e+02 3.266000e+02
## resY.mean.log1p 6.185647e+00 6.186572e+00 5.791793e+00
## resY.mean.nexp 1.443518e-142 2.530221e-127 3.059287e-211
## resY.mean.root2 2.201652e+01 2.202672e+01 1.807208e+01
## resY.min 3.750000e+02 3.750000e+02 5.000000e+01
## resY.min.log1p 5.929589e+00 5.929589e+00 3.931826e+00
## resY.min.nexp 1.928750e-22 7.095474e-23 1.379016e-163
## resY.min.root2 1.936492e+01 1.936492e+01 7.071068e+00
## min.outdoor.fctr.Y max.outdoor.fctr.All.X..rcv.glmnet.Y
## .pos 1.500000e+01 1.999000e+03
## resX.mean 3.475000e+02 5.000000e+02
## resX.mean.log1p 5.853638e+00 6.216606e+00
## resX.mean.nexp 2.874259e-215 5.762208e-124
## resX.mean.root2 1.864135e+01 2.236068e+01
## resX.min 6.800000e+01 5.000000e+02
## resX.min.log1p 4.234107e+00 6.216606e+00
## resX.min.root2 8.246211e+00 2.236068e+01
## resXY.mean 1.137250e+05 2.182778e+05
## resXY.mean.log1p 1.164155e+01 1.229353e+01
## resXY.mean.root2 3.372314e+02 4.672021e+02
## resY.mad 0.000000e+00 1.942206e+02
## resY.mad.log1p 0.000000e+00 5.274130e+00
## resY.mad.root2 0.000000e+00 1.393631e+01
## resY.mean 2.915000e+02 5.000000e+02
## resY.mean.log1p 5.678465e+00 6.216606e+00
## resY.mean.nexp 1.952253e-211 1.328912e-110
## resY.mean.root2 1.707337e+01 2.236068e+01
## resY.min 5.100000e+01 5.000000e+02
## resY.min.log1p 3.951244e+00 6.216606e+00
## resY.min.nexp 1.379016e-163 2.543666e-13
## resY.min.root2 7.141428e+00 2.236068e+01
## min.outdoor.fctr.All.X..rcv.glmnet.Y
## .pos 1.000000e+00
## resX.mean 2.837692e+02
## resX.mean.log1p 5.651679e+00
## resX.mean.nexp 7.124576e-218
## resX.mean.root2 1.684545e+01
## resX.min 6.800000e+01
## resX.min.log1p 4.234107e+00
## resX.min.root2 8.246211e+00
## resXY.mean 8.762615e+04
## resXY.mean.log1p 1.138085e+01
## resXY.mean.root2 2.960172e+02
## resY.mad 0.000000e+00
## resY.mad.log1p 0.000000e+00
## resY.mad.root2 0.000000e+00
## resY.mean 2.530000e+02
## resY.mean.log1p 5.537334e+00
## resY.mean.nexp 7.124576e-218
## resY.mean.root2 1.590597e+01
## resY.min 2.900000e+01
## resY.min.log1p 3.401197e+00
## resY.min.nexp 7.124576e-218
## resY.min.root2 5.385165e+00
## max.outdoor.fctr.Final..rcv.glmnet.Y
## .pos 1.200000e+04
## resX.mean 5.000000e+02
## resX.mean.log1p 6.216606e+00
## resX.mean.nexp 5.719134e-133
## resX.mean.root2 2.236068e+01
## resX.min 5.000000e+02
## resX.min.log1p 6.216606e+00
## resX.min.root2 2.236068e+01
## resXY.mean 2.500000e+05
## resXY.mean.log1p 1.242922e+01
## resXY.mean.root2 5.000000e+02
## resY.mad 2.787288e+02
## resY.mad.log1p 5.633821e+00
## resY.mad.root2 1.669517e+01
## resY.mean 5.000000e+02
## resY.mean.log1p 6.216606e+00
## resY.mean.nexp 1.105028e-116
## resY.mean.root2 2.236068e+01
## resY.min 5.000000e+02
## resY.min.log1p 6.216606e+00
## resY.min.nexp 2.543666e-13
## resY.min.root2 2.236068e+01
## min.outdoor.fctr.Final..rcv.glmnet.Y
## .pos 2.001000e+03
## resX.mean 3.045000e+02
## resX.mean.log1p 5.721950e+00
## resX.mean.nexp 7.124576e-218
## resX.mean.root2 1.744993e+01
## resX.min 5.300000e+01
## resX.min.log1p 3.988984e+00
## resX.min.root2 7.280110e+00
## resXY.mean 1.058460e+05
## resXY.mean.log1p 1.156975e+01
## resXY.mean.root2 3.253398e+02
## resY.mad 0.000000e+00
## resY.mad.log1p 0.000000e+00
## resY.mad.root2 0.000000e+00
## resY.mean 2.670000e+02
## resY.mean.log1p 5.590987e+00
## resY.mean.nexp 7.124576e-218
## resY.mean.root2 1.634013e+01
## resY.min 2.900000e+01
## resY.min.log1p 3.401197e+00
## resY.min.nexp 7.124576e-218
## resY.min.root2 5.385165e+00
## [1] "OOBobs total range outliers: 38"
## [1] "newobs outdoor.fctr.Final..rcv.glmnet Y: min < min of Train range: 43"
## business_id outdoor.fctr.Final..rcv.glmnet nImgs nImgs.log1p
## 2222 0qj4g Y 12 2.5649494
## 2316 12p62 Y 1 0.6931472
## 2363 18cak Y 15 2.7725887
## 2473 1lvgn Y 1 0.6931472
## 2821 2yeud Y 96 4.5747110
## 2850 31y16 Y 6 1.9459101
## nImgs.root2 resX.mean resX.mean.log1p resX.mean.root2 resX.min
## 2222 3.464102 382.1667 5.948470 19.54908 53
## 2316 1.000000 500.0000 6.216606 22.36068 500
## 2363 3.872983 343.2667 5.841417 18.52746 110
## 2473 1.000000 500.0000 6.216606 22.36068 500
## 2821 9.797959 451.9271 6.115731 21.25858 280
## 2850 2.449490 344.5000 5.844993 18.56071 96
## resX.min.log1p resX.min.root2 resXY.mean resXY.mean.log1p
## 2222 3.988984 7.280110 171250.5 12.05089
## 2316 6.216606 22.360680 187500.0 12.14154
## 2363 4.709530 10.488088 122284.5 11.71411
## 2473 6.216606 22.360680 187500.0 12.14154
## 2821 5.638355 16.733201 183807.3 12.12165
## 2850 4.574711 9.797959 105846.0 11.56975
## resXY.mean.root2 resXY.min resXY.min.log1p resXY.min.root2
## 2222 413.8242 1802 7.497207 42.44997
## 2316 433.0127 187500 12.141539 433.01270
## 2363 349.6920 28160 10.245693 167.80942
## 2473 433.0127 187500 12.141539 433.01270
## 2821 428.7275 20500 9.928229 143.17821
## 2850 325.3398 12288 9.416460 110.85125
## resY.mad.nexp resY.mean resY.mean.log1p resY.mean.root2 resY.min
## 2222 1.000000e+00 414.3333 6.029081 20.35518 34
## 2316 1.000000e+00 375.0000 5.929589 19.36492 375
## 2363 2.635452e-50 341.2000 5.835395 18.47160 198
## 2473 1.000000e+00 375.0000 5.929589 19.36492 375
## 2821 1.000000e+00 415.6875 6.032337 20.38842 41
## 2850 2.635452e-50 267.0000 5.590987 16.34013 128
## resY.min.log1p resY.min.root2
## 2222 3.555348 5.830952
## 2316 5.929589 19.364917
## 2363 5.293305 14.071247
## 2473 5.929589 19.364917
## 2821 3.737670 6.403124
## 2850 4.859812 11.313708
## business_id outdoor.fctr.Final..rcv.glmnet nImgs nImgs.log1p
## 2363 18cak Y 15 2.772589
## 3310 4q512 Y 2 1.098612
## 5530 cijo0 Y 34 3.555348
## 6364 fjvig Y 578 6.361302
## 9792 rx5mm Y 867 6.766192
## 11906 znmff Y 17 2.890372
## nImgs.root2 resX.mean resX.mean.log1p resX.mean.root2 resX.min
## 2363 3.872983 343.2667 5.841417 18.52746 110
## 3310 1.414214 304.5000 5.721950 17.44993 234
## 5530 5.830952 426.2353 6.057335 20.64547 281
## 6364 24.041631 439.3478 6.087565 20.96062 63
## 9792 29.444864 439.4464 6.087789 20.96298 63
## 11906 4.123106 344.2941 5.844397 18.55516 97
## resX.min.log1p resX.min.root2 resXY.mean resXY.mean.log1p
## 2363 4.709530 10.488088 122284.5 11.71411
## 3310 5.459586 15.297059 114225.0 11.64593
## 5530 5.641907 16.763055 174044.1 12.06707
## 6364 4.158883 7.937254 185841.0 12.13265
## 9792 4.158883 7.937254 185129.7 12.12882
## 11906 4.584967 9.848858 127450.6 11.75549
## resXY.mean.root2 resXY.min resXY.min.log1p resXY.min.root2
## 2363 349.6920 28160 10.245693 167.80942
## 3310 337.9719 40950 10.620132 202.36106
## 5530 417.1860 20500 9.928229 143.17821
## 6364 431.0928 9204 9.127502 95.93748
## 9792 430.2670 10836 9.290721 104.09611
## 11906 357.0022 12610 9.442325 112.29426
## resY.mad.nexp resY.mean resY.mean.log1p resY.mean.root2 resY.min
## 2363 2.635452e-50 341.2000 5.835395 18.47160 198
## 3310 2.337156e-105 337.5000 5.824524 18.37117 175
## 5530 1.000000e+00 421.8529 6.047024 20.53906 41
## 6364 1.000000e+00 429.4844 6.064911 20.72401 78
## 9792 1.000000e+00 429.3633 6.064630 20.72108 156
## 11906 3.268701e-81 305.3529 5.724738 17.47435 97
## resY.min.log1p resY.min.root2
## 2363 5.293305 14.071247
## 3310 5.170484 13.228757
## 5530 3.737670 6.403124
## 6364 4.369448 8.831761
## 9792 5.056246 12.489996
## 11906 4.584967 9.848858
## business_id outdoor.fctr.Final..rcv.glmnet nImgs nImgs.log1p
## 10697 v7cb4 Y 741 6.609349
## 11700 yww34 Y 598 6.395262
## 11824 zedxo Y 1962 7.582229
## 11838 zfrmk Y 89 4.499810
## 11906 znmff Y 17 2.890372
## 11942 zsd8q Y 31 3.465736
## nImgs.root2 resX.mean resX.mean.log1p resX.mean.root2 resX.min
## 10697 27.221315 442.8421 6.095469 21.04381 63
## 11700 24.454039 440.5217 6.090227 20.98861 67
## 11824 44.294469 439.5061 6.087924 20.96440 63
## 11838 9.433981 429.3708 6.064647 20.72126 63
## 11906 4.123106 344.2941 5.844397 18.55516 97
## 11942 5.567764 434.0323 6.075420 20.83344 63
## resX.min.log1p resX.min.root2 resXY.mean resXY.mean.log1p
## 10697 4.158883 7.937254 186260.2 12.13491
## 11700 4.219508 8.185353 184358.9 12.12465
## 11824 4.158883 7.937254 183954.9 12.12245
## 11838 4.158883 7.937254 180279.1 12.10227
## 11906 4.584967 9.848858 127450.6 11.75549
## 11942 4.158883 7.937254 165092.0 12.01426
## resXY.mean.root2 resXY.min resXY.min.log1p resXY.min.root2
## 10697 431.5787 10836 9.290721 104.09611
## 11700 429.3704 6700 8.810012 81.85353
## 11824 428.8996 10836 9.290721 104.09611
## 11838 424.5928 10836 9.290721 104.09611
## 11906 357.0022 12610 9.442325 112.29426
## 11942 406.3152 10836 9.290721 104.09611
## resY.mad.nexp resY.mean resY.mean.log1p resY.mean.root2 resY.min
## 10697 2.466466e-68 428.2740 6.062095 20.69478 160
## 11700 1.421895e-61 426.6338 6.058267 20.65512 77
## 11824 1.623407e-25 426.8282 6.058722 20.65982 84
## 11838 1.603375e-06 427.9663 6.061378 20.68735 172
## 11906 3.268701e-81 305.3529 5.724738 17.47435 97
## 11942 3.987632e-27 377.2258 5.935491 19.42230 172
## resY.min.log1p resY.min.root2
## 10697 5.081404 12.649111
## 11700 4.356709 8.774964
## 11824 4.442651 9.165151
## 11838 5.153292 13.114877
## 11906 4.584967 9.848858
## 11942 5.153292 13.114877
## id cor.y exclude.as.feat cor.y.abs
## nImgs nImgs -0.014963676 FALSE 0.014963676
## nImgs.log1p nImgs.log1p 0.047250893 FALSE 0.047250893
## nImgs.root2 nImgs.root2 0.014028124 FALSE 0.014028124
## resX.mean resX.mean -0.017726551 FALSE 0.017726551
## resX.mean.log1p resX.mean.log1p -0.015059015 FALSE 0.015059015
## resX.mean.root2 resX.mean.root2 -0.016434019 FALSE 0.016434019
## resX.min resX.min -0.031436275 FALSE 0.031436275
## resX.min.log1p resX.min.log1p -0.030103276 FALSE 0.030103276
## resX.min.root2 resX.min.root2 -0.030339745 FALSE 0.030339745
## resXY.mean resXY.mean -0.009002880 FALSE 0.009002880
## resXY.mean.log1p resXY.mean.log1p -0.004867571 FALSE 0.004867571
## resXY.mean.root2 resXY.mean.root2 -0.007039955 FALSE 0.007039955
## resXY.min resXY.min -0.049458217 FALSE 0.049458217
## resXY.min.log1p resXY.min.log1p -0.033756424 FALSE 0.033756424
## resXY.min.root2 resXY.min.root2 -0.041449898 FALSE 0.041449898
## resY.mad.nexp resY.mad.nexp 0.012190340 FALSE 0.012190340
## resY.mean resY.mean 0.012599188 FALSE 0.012599188
## resY.mean.log1p resY.mean.log1p 0.013625190 FALSE 0.013625190
## resY.mean.root2 resY.mean.root2 0.013106506 FALSE 0.013106506
## resY.min resY.min -0.050925308 FALSE 0.050925308
## resY.min.log1p resY.min.log1p -0.043072548 FALSE 0.043072548
## resY.min.root2 resY.min.root2 -0.047387777 FALSE 0.047387777
## cor.high.X freqRatio percentUnique zeroVar nzv
## nImgs <NA> 1.033333 19.10 FALSE FALSE
## nImgs.log1p nImgs.cut.fctr 1.033333 19.10 FALSE FALSE
## nImgs.root2 nImgs.log1p 1.033333 19.10 FALSE FALSE
## resX.mean <NA> 2.000000 97.75 FALSE FALSE
## resX.mean.log1p resX.mean 2.000000 97.60 FALSE FALSE
## resX.mean.root2 resX.mean 2.000000 97.45 FALSE FALSE
## resX.min <NA> 6.000000 11.45 FALSE FALSE
## resX.min.log1p resX.min 6.000000 11.45 FALSE FALSE
## resX.min.root2 resX.min 6.000000 11.45 FALSE FALSE
## resXY.mean <NA> 6.000000 98.55 FALSE FALSE
## resXY.mean.log1p <NA> 4.000000 90.80 FALSE FALSE
## resXY.mean.root2 <NA> 6.000000 98.20 FALSE FALSE
## resXY.min resY.min 9.745455 37.65 FALSE FALSE
## resXY.min.log1p resXY.min 9.745455 37.65 FALSE FALSE
## resXY.min.root2 resXY.min 9.745455 37.65 FALSE FALSE
## resY.mad.nexp <NA> 5.354497 9.05 FALSE FALSE
## resY.mean resY.mean.root2 1.666667 98.15 FALSE FALSE
## resY.mean.log1p <NA> 1.666667 97.90 FALSE FALSE
## resY.mean.root2 resY.mean.log1p 1.666667 97.85 FALSE FALSE
## resY.min <NA> 9.824561 13.85 FALSE FALSE
## resY.min.log1p resY.min 9.824561 13.85 FALSE FALSE
## resY.min.root2 resY.min 9.824561 13.85 FALSE FALSE
## is.cor.y.abs.low interaction.feat shapiro.test.p.value
## nImgs FALSE NA 1.364097e-61
## nImgs.log1p FALSE NA 1.234907e-13
## nImgs.root2 FALSE NA 4.118632e-46
## resX.mean FALSE NA 1.161337e-19
## resX.mean.log1p FALSE NA 2.973500e-25
## resX.mean.root2 FALSE NA 1.959497e-22
## resX.min FALSE NA 2.978480e-35
## resX.min.log1p FALSE NA 1.543786e-43
## resX.min.root2 FALSE NA 7.580680e-39
## resXY.mean FALSE NA 2.964553e-36
## resXY.mean.log1p TRUE NA 6.980019e-43
## resXY.mean.root2 TRUE NA 1.780045e-39
## resXY.min FALSE NA 2.084930e-32
## resXY.min.log1p FALSE NA 1.076069e-42
## resXY.min.root2 FALSE NA 1.752753e-35
## resY.mad.nexp FALSE NA 1.839563e-53
## resY.mean FALSE NA 1.464051e-21
## resY.mean.log1p FALSE NA 3.854130e-28
## resY.mean.root2 FALSE NA 7.216658e-25
## resY.min FALSE NA 1.973528e-28
## resY.min.log1p FALSE NA 3.088017e-38
## resY.min.root2 FALSE NA 2.137435e-32
## rsp_var_raw id_var rsp_var max min
## nImgs FALSE NA NA 2.974000e+03 1.000000e+00
## nImgs.log1p FALSE NA NA 7.997999e+00 6.931472e-01
## nImgs.root2 FALSE NA NA 5.453439e+01 1.000000e+00
## resX.mean FALSE NA NA 5.000000e+02 2.837692e+02
## resX.mean.log1p FALSE NA NA 6.216606e+00 5.651679e+00
## resX.mean.root2 FALSE NA NA 2.236068e+01 1.684545e+01
## resX.min FALSE NA NA 5.000000e+02 5.300000e+01
## resX.min.log1p FALSE NA NA 6.216606e+00 3.988984e+00
## resX.min.root2 FALSE NA NA 2.236068e+01 7.280110e+00
## resXY.mean FALSE NA NA 2.500000e+05 8.762615e+04
## resXY.mean.log1p FALSE NA NA 1.242922e+01 1.138085e+01
## resXY.mean.root2 FALSE NA NA 5.000000e+02 2.960172e+02
## resXY.min FALSE NA NA 2.500000e+05 1.802000e+03
## resXY.min.log1p FALSE NA NA 1.242922e+01 7.497207e+00
## resXY.min.root2 FALSE NA NA 5.000000e+02 4.244997e+01
## resY.mad.nexp FALSE NA NA 1.000000e+00 8.904719e-122
## resY.mean FALSE NA NA 5.000000e+02 2.530000e+02
## resY.mean.log1p FALSE NA NA 6.216606e+00 5.537334e+00
## resY.mean.root2 FALSE NA NA 2.236068e+01 1.590597e+01
## resY.min FALSE NA NA 5.000000e+02 2.900000e+01
## resY.min.log1p FALSE NA NA 6.216606e+00 3.401197e+00
## resY.min.root2 FALSE NA NA 2.236068e+01 5.385165e+00
## max.outdoor.fctr.N max.outdoor.fctr.Y min.outdoor.fctr.N
## nImgs 2.974000e+03 1.954000e+03 2.000000e+00
## nImgs.log1p 7.997999e+00 7.578145e+00 1.098612e+00
## nImgs.root2 5.453439e+01 4.420407e+01 1.414214e+00
## resX.mean 5.000000e+02 5.000000e+02 2.837692e+02
## resX.mean.log1p 6.216606e+00 6.216606e+00 5.651679e+00
## resX.mean.root2 2.236068e+01 2.236068e+01 1.684545e+01
## resX.min 5.000000e+02 5.000000e+02 6.100000e+01
## resX.min.log1p 6.216606e+00 6.216606e+00 4.127134e+00
## resX.min.root2 2.236068e+01 2.236068e+01 7.810250e+00
## resXY.mean 2.175740e+05 2.182778e+05 8.762615e+04
## resXY.mean.log1p 1.229030e+01 1.229353e+01 1.138085e+01
## resXY.mean.root2 4.664483e+02 4.672021e+02 2.960172e+02
## resXY.min 1.875000e+05 1.875000e+05 5.000000e+03
## resXY.min.log1p 1.214154e+01 1.214154e+01 8.517393e+00
## resXY.min.root2 4.330127e+02 4.330127e+02 7.071068e+01
## resY.mad.nexp 1.000000e+00 1.000000e+00 4.477805e-85
## resY.mean 4.861111e+02 5.000000e+02 2.530000e+02
## resY.mean.log1p 6.188492e+00 6.216606e+00 5.537334e+00
## resY.mean.root2 2.204793e+01 2.236068e+01 1.590597e+01
## resY.min 3.750000e+02 5.000000e+02 2.900000e+01
## resY.min.log1p 5.929589e+00 6.216606e+00 3.401197e+00
## resY.min.root2 1.936492e+01 2.236068e+01 5.385165e+00
## min.outdoor.fctr.Y max.outdoor.fctr.All.X..rcv.glmnet.Y
## nImgs 2.000000e+00 1.750000e+03
## nImgs.log1p 1.098612e+00 7.467942e+00
## nImgs.root2 1.414214e+00 4.183300e+01
## resX.mean 3.475000e+02 5.000000e+02
## resX.mean.log1p 5.853638e+00 6.216606e+00
## resX.mean.root2 1.864135e+01 2.236068e+01
## resX.min 6.800000e+01 5.000000e+02
## resX.min.log1p 4.234107e+00 6.216606e+00
## resX.min.root2 8.246211e+00 2.236068e+01
## resXY.mean 1.137250e+05 2.182778e+05
## resXY.mean.log1p 1.164155e+01 1.229353e+01
## resXY.mean.root2 3.372314e+02 4.672021e+02
## resXY.min 3.675000e+03 1.875000e+05
## resXY.min.log1p 8.209580e+00 1.214154e+01
## resXY.min.root2 6.062178e+01 4.330127e+02
## resY.mad.nexp 3.268701e-81 1.000000e+00
## resY.mean 2.915000e+02 5.000000e+02
## resY.mean.log1p 5.678465e+00 6.216606e+00
## resY.mean.root2 1.707337e+01 2.236068e+01
## resY.min 4.400000e+01 5.000000e+02
## resY.min.log1p 3.806662e+00 6.216606e+00
## resY.min.root2 6.633250e+00 2.236068e+01
## min.outdoor.fctr.All.X..rcv.glmnet.Y
## nImgs 2.000000e+00
## nImgs.log1p 1.098612e+00
## nImgs.root2 1.414214e+00
## resX.mean 2.837692e+02
## resX.mean.log1p 5.651679e+00
## resX.mean.root2 1.684545e+01
## resX.min 6.800000e+01
## resX.min.log1p 4.234107e+00
## resX.min.root2 8.246211e+00
## resXY.mean 8.762615e+04
## resXY.mean.log1p 1.138085e+01
## resXY.mean.root2 2.960172e+02
## resXY.min 3.675000e+03
## resXY.min.log1p 8.209580e+00
## resXY.min.root2 6.062178e+01
## resY.mad.nexp 4.477805e-85
## resY.mean 2.530000e+02
## resY.mean.log1p 5.537334e+00
## resY.mean.root2 1.590597e+01
## resY.min 2.900000e+01
## resY.min.log1p 3.401197e+00
## resY.min.root2 5.385165e+00
## max.outdoor.fctr.Final..rcv.glmnet.Y
## nImgs 2.825000e+03
## nImgs.log1p 7.946618e+00
## nImgs.root2 5.315073e+01
## resX.mean 5.000000e+02
## resX.mean.log1p 6.216606e+00
## resX.mean.root2 2.236068e+01
## resX.min 5.000000e+02
## resX.min.log1p 6.216606e+00
## resX.min.root2 2.236068e+01
## resXY.mean 2.500000e+05
## resXY.mean.log1p 1.242922e+01
## resXY.mean.root2 5.000000e+02
## resXY.min 2.500000e+05
## resXY.min.log1p 1.242922e+01
## resXY.min.root2 5.000000e+02
## resY.mad.nexp 1.000000e+00
## resY.mean 5.000000e+02
## resY.mean.log1p 6.216606e+00
## resY.mean.root2 2.236068e+01
## resY.min 5.000000e+02
## resY.min.log1p 6.216606e+00
## resY.min.root2 2.236068e+01
## min.outdoor.fctr.Final..rcv.glmnet.Y
## nImgs 1.000000e+00
## nImgs.log1p 6.931472e-01
## nImgs.root2 1.000000e+00
## resX.mean 3.045000e+02
## resX.mean.log1p 5.721950e+00
## resX.mean.root2 1.744993e+01
## resX.min 5.300000e+01
## resX.min.log1p 3.988984e+00
## resX.min.root2 7.280110e+00
## resXY.mean 1.058460e+05
## resXY.mean.log1p 1.156975e+01
## resXY.mean.root2 3.253398e+02
## resXY.min 1.802000e+03
## resXY.min.log1p 7.497207e+00
## resXY.min.root2 4.244997e+01
## resY.mad.nexp 8.904719e-122
## resY.mean 2.670000e+02
## resY.mean.log1p 5.590987e+00
## resY.mean.root2 1.634013e+01
## resY.min 2.900000e+01
## resY.min.log1p 3.401197e+00
## resY.min.root2 5.385165e+00
## [1] "newobs outdoor.fctr.Final..rcv.glmnet Y: max > max of Train range: 10000"
## business_id outdoor.fctr.Final..rcv.glmnet .pos nImgs nImgs.log1p
## 2001 003sg Y 2001 167 5.123964
## 2002 00er5 Y 2002 210 5.351858
## 2003 00kad Y 2003 83 4.430817
## 2004 00mc6 Y 2004 15 2.772589
## 2005 00q7x Y 2005 24 3.218876
## 2006 00v0t Y 2006 24 3.218876
## nImgs.nexp nImgs.root2 resX.mean.nexp resX.min.nexp resXY.mad
## 2001 2.970445e-73 12.922848 3.225211e-191 1.294998e-100 0
## 2002 6.282881e-92 14.491377 2.758174e-188 9.188626e-123 0
## 2003 8.985826e-37 9.110434 5.050852e-187 9.188626e-123 0
## 2004 3.059023e-07 3.872983 6.231837e-192 9.188626e-123 0
## 2005 3.775135e-11 4.898979 2.589612e-194 7.149792e-142 0
## 2006 3.775135e-11 4.898979 3.224442e-195 1.018963e-162 0
## resXY.mad.log1p resXY.mad.root2 resXY.mean resXY.mean.log1p
## 2001 0 0 185382.6 12.13018
## 2002 0 0 181563.1 12.10936
## 2003 0 0 182564.8 12.11487
## 2004 0 0 191000.0 12.16003
## 2005 0 0 190645.8 12.15818
## 2006 0 0 188770.8 12.14829
## resXY.mean.root2 resXY.min resXY.min.log1p resXY.min.root2 resY.mad
## 2001 430.5608 92050 11.43010 303.3974 0.0000
## 2002 426.1022 65880 11.09561 256.6710 0.0000
## 2003 427.2760 93375 11.44439 305.5732 0.0000
## 2004 437.0355 140500 11.85297 374.8333 0.0000
## 2005 436.6301 162500 11.99844 403.1129 92.6625
## 2006 434.4777 140500 11.85297 374.8333 92.6625
## resY.mad.log1p resY.mad.root2 resY.mean.nexp resY.min.nexp
## 2001 0.000000 0.000000 4.335495e-188 5.583037e-85
## 2002 0.000000 0.000000 8.526171e-188 3.342796e-80
## 2003 0.000000 0.000000 8.016718e-190 7.255611e-109
## 2004 0.000000 0.000000 1.437642e-192 2.398488e-145
## 2005 4.539698 9.626136 7.025181e-190 2.398488e-145
## 2006 4.539698 9.626136 2.399065e-187 9.188626e-123
## business_id outdoor.fctr.Final..rcv.glmnet .pos nImgs nImgs.log1p
## 3111 4051c Y 3111 61 4.127134
## 5473 cb1v8 Y 5473 34 3.555348
## 6175 ewq0g Y 6175 58 4.077537
## 8809 oftao Y 8809 32 3.496508
## 9248 pzr3p Y 9248 298 5.700444
## 11902 zn3oa Y 11902 56 4.043051
## nImgs.nexp nImgs.root2 resX.mean.nexp resX.min.nexp resXY.mad
## 3111 3.221340e-27 7.810250 1.615271e-195 9.188626e-123 0.0
## 5473 1.713908e-15 5.830952 7.900500e-192 3.481107e-57 741.3
## 6175 6.470235e-26 7.615773 5.356061e-197 9.188626e-123 0.0
## 8809 1.266417e-14 5.656854 2.694379e-190 9.188626e-123 0.0
## 9248 3.804034e-130 17.262677 9.404655e-193 3.706956e-120 0.0
## 11902 4.780893e-25 7.483315 5.059170e-203 3.082441e-126 0.0
## resXY.mad.log1p resXY.mad.root2 resXY.mean resXY.mean.log1p
## 3111 0.000000 0.00000 186573.8 12.13659
## 5473 6.609753 27.22683 186521.8 12.13631
## 6175 0.000000 0.00000 186617.2 12.13682
## 8809 0.000000 0.00000 182203.1 12.11288
## 9248 0.000000 0.00000 183746.0 12.12131
## 11902 0.000000 0.00000 190526.8 12.15755
## resXY.mean.root2 resXY.min resXY.min.log1p resXY.min.root2 resY.mad
## 3111 431.9419 140500 11.85297 374.8333 139.3644
## 5473 431.8817 12740 9.45258 112.8716 11.1195
## 6175 431.9922 76800 11.24897 277.1281 114.1602
## 8809 426.8526 140500 11.85297 374.8333 92.6625
## 9248 428.6560 43500 10.68054 208.5665 139.3644
## 11902 436.4937 144500 11.88104 380.1316 0.0000
## resY.mad.log1p resY.mad.root2 resY.mean.nexp resY.min.nexp
## 3111 4.944242 11.805270 3.877891e-185 9.188626e-123
## 5473 2.494816 3.334591 5.548548e-185 2.748785e-43
## 6175 4.746324 10.684578 5.383921e-183 5.879283e-105
## 8809 4.539698 9.626136 1.454377e-186 9.188626e-123
## 9248 4.944242 11.805270 2.615522e-185 1.645811e-38
## 11902 0.000000 0.000000 4.562642e-181 3.082441e-126
## business_id outdoor.fctr.Final..rcv.glmnet .pos nImgs nImgs.log1p
## 11995 zyrif Y 11995 89 4.499810
## 11996 zyvg6 Y 11996 16 2.833213
## 11997 zyvjj Y 11997 27 3.332205
## 11998 zz8g4 Y 11998 118 4.779123
## 11999 zzxkg Y 11999 154 5.043425
## 12000 zzxwm Y 12000 13 2.639057
## nImgs.nexp nImgs.root2 resX.mean.nexp resX.min.nexp resXY.mad
## 11995 2.227364e-39 9.433981 2.552892e-180 1.076828e-106 1482.6
## 11996 1.125352e-07 4.000000 4.256714e-188 2.497728e-122 0.0
## 11997 1.879529e-12 5.196152 1.115844e-200 9.188626e-123 1482.6
## 11998 5.665668e-52 10.862780 5.049362e-193 9.188626e-123 0.0
## 11999 1.314165e-67 12.409674 1.561603e-190 9.188626e-123 0.0
## 12000 2.260329e-06 3.605551 8.694857e-179 2.107672e-95 1482.6
## resXY.mad.log1p resXY.mad.root2 resXY.mean resXY.mean.log1p
## 11995 7.302227 38.50455 175542.2 12.07564
## 11996 0.000000 0.00000 188375.0 12.14620
## 11997 7.302227 38.50455 179504.8 12.09796
## 11998 0.000000 0.00000 183586.2 12.12045
## 11999 0.000000 0.00000 182534.7 12.11470
## 12000 7.302227 38.50455 161115.2 11.98988
## resXY.mean.root2 resXY.min resXY.min.log1p resXY.min.root2 resY.mad
## 11995 418.9776 99552 11.50845 315.5186 0.0000
## 11996 434.0219 140000 11.84940 374.1657 0.0000
## 11997 423.6801 126630 11.74903 355.8511 2.9652
## 11998 428.4697 116000 11.66135 340.5877 169.0164
## 11999 427.2408 76800 11.24897 277.1281 106.7472
## 12000 401.3916 45998 10.73637 214.4714 185.3250
## resY.mad.log1p resY.mad.root2 resY.mean.nexp resY.min.nexp
## 11995 0.000000 0.000000 1.321107e-189 9.188626e-123
## 11996 0.000000 0.000000 4.010869e-194 1.379016e-163
## 11997 1.377556 1.721976 2.968218e-173 2.497728e-122
## 11998 5.135895 13.000631 4.895966e-185 1.752589e-101
## 11999 4.679788 10.331854 5.383963e-186 5.879283e-105
## 12000 5.227492 13.613413 2.605171e-174 2.311343e-92
## id cor.y exclude.as.feat cor.y.abs
## .pos .pos 0.027497300 FALSE 0.027497300
## nImgs nImgs -0.014963676 FALSE 0.014963676
## nImgs.log1p nImgs.log1p 0.047250893 FALSE 0.047250893
## nImgs.nexp nImgs.nexp -0.003435316 FALSE 0.003435316
## nImgs.root2 nImgs.root2 0.014028124 FALSE 0.014028124
## resX.mean.nexp resX.mean.nexp -0.022433472 FALSE 0.022433472
## resX.min.nexp resX.min.nexp -0.022391602 FALSE 0.022391602
## resXY.mad resXY.mad -0.011946049 FALSE 0.011946049
## resXY.mad.log1p resXY.mad.log1p -0.014055066 FALSE 0.014055066
## resXY.mad.root2 resXY.mad.root2 -0.011364822 FALSE 0.011364822
## resXY.mean resXY.mean -0.009002880 FALSE 0.009002880
## resXY.mean.log1p resXY.mean.log1p -0.004867571 FALSE 0.004867571
## resXY.mean.root2 resXY.mean.root2 -0.007039955 FALSE 0.007039955
## resXY.min resXY.min -0.049458217 FALSE 0.049458217
## resXY.min.log1p resXY.min.log1p -0.033756424 FALSE 0.033756424
## resXY.min.root2 resXY.min.root2 -0.041449898 FALSE 0.041449898
## resY.mad resY.mad 0.007630633 FALSE 0.007630633
## resY.mad.log1p resY.mad.log1p -0.001526058 FALSE 0.001526058
## resY.mad.root2 resY.mad.root2 0.002557583 FALSE 0.002557583
## resY.mean.nexp resY.mean.nexp -0.022433472 FALSE 0.022433472
## resY.min.nexp resY.min.nexp -0.022433600 FALSE 0.022433600
## cor.high.X freqRatio percentUnique zeroVar nzv
## .pos <NA> 1.000000 100.00 FALSE FALSE
## nImgs <NA> 1.033333 19.10 FALSE FALSE
## nImgs.log1p nImgs.cut.fctr 1.033333 19.10 FALSE FALSE
## nImgs.nexp <NA> 1.193548 17.35 FALSE FALSE
## nImgs.root2 nImgs.log1p 1.033333 19.10 FALSE FALSE
## resX.mean.nexp <NA> 2.000000 97.75 FALSE FALSE
## resX.min.nexp <NA> 6.000000 11.45 FALSE FALSE
## resXY.mad <NA> 9.568047 4.35 FALSE FALSE
## resXY.mad.log1p resXY.mad.nexp 9.568047 4.35 FALSE FALSE
## resXY.mad.root2 resXY.mad 9.568047 4.35 FALSE FALSE
## resXY.mean <NA> 6.000000 98.55 FALSE FALSE
## resXY.mean.log1p <NA> 4.000000 90.80 FALSE FALSE
## resXY.mean.root2 <NA> 6.000000 98.20 FALSE FALSE
## resXY.min resY.min 9.745455 37.65 FALSE FALSE
## resXY.min.log1p resXY.min 9.745455 37.65 FALSE FALSE
## resXY.min.root2 resXY.min 9.745455 37.65 FALSE FALSE
## resY.mad <NA> 5.354497 9.05 FALSE FALSE
## resY.mad.log1p <NA> 5.354497 9.05 FALSE FALSE
## resY.mad.root2 <NA> 5.354497 9.05 FALSE FALSE
## resY.mean.nexp resX.mean.nexp 1.666667 98.15 FALSE FALSE
## resY.min.nexp <NA> 9.824561 13.85 FALSE FALSE
## is.cor.y.abs.low interaction.feat shapiro.test.p.value
## .pos FALSE NA 2.145811e-24
## nImgs FALSE NA 1.364097e-61
## nImgs.log1p FALSE NA 1.234907e-13
## nImgs.nexp TRUE NA 1.763177e-72
## nImgs.root2 FALSE NA 4.118632e-46
## resX.mean.nexp FALSE NA 1.194234e-72
## resX.min.nexp FALSE NA 1.195403e-72
## resXY.mad FALSE NA 9.894151e-67
## resXY.mad.log1p FALSE NA 3.868763e-59
## resXY.mad.root2 FALSE NA 1.509232e-62
## resXY.mean FALSE NA 2.964553e-36
## resXY.mean.log1p TRUE NA 6.980019e-43
## resXY.mean.root2 TRUE NA 1.780045e-39
## resXY.min FALSE NA 2.084930e-32
## resXY.min.log1p FALSE NA 1.076069e-42
## resXY.min.root2 FALSE NA 1.752753e-35
## resY.mad TRUE NA 3.711302e-48
## resY.mad.log1p TRUE NA 3.133148e-49
## resY.mad.root2 TRUE NA 1.717662e-47
## resY.mean.nexp FALSE NA 1.194234e-72
## resY.min.nexp FALSE NA 1.194238e-72
## rsp_var_raw id_var rsp_var max min
## .pos FALSE NA NA 1.200000e+04 1.000000e+00
## nImgs FALSE NA NA 2.974000e+03 1.000000e+00
## nImgs.log1p FALSE NA NA 7.997999e+00 6.931472e-01
## nImgs.nexp FALSE NA NA 3.678794e-01 0.000000e+00
## nImgs.root2 FALSE NA NA 5.453439e+01 1.000000e+00
## resX.mean.nexp FALSE NA NA 5.762208e-124 7.124576e-218
## resX.min.nexp FALSE NA NA 9.602680e-24 7.124576e-218
## resXY.mad FALSE NA NA 1.237971e+05 0.000000e+00
## resXY.mad.log1p FALSE NA NA 1.172641e+01 0.000000e+00
## resXY.mad.root2 FALSE NA NA 3.518481e+02 0.000000e+00
## resXY.mean FALSE NA NA 2.500000e+05 8.762615e+04
## resXY.mean.log1p FALSE NA NA 1.242922e+01 1.138085e+01
## resXY.mean.root2 FALSE NA NA 5.000000e+02 2.960172e+02
## resXY.min FALSE NA NA 2.500000e+05 1.802000e+03
## resXY.min.log1p FALSE NA NA 1.242922e+01 7.497207e+00
## resXY.min.root2 FALSE NA NA 5.000000e+02 4.244997e+01
## resY.mad FALSE NA NA 2.787288e+02 0.000000e+00
## resY.mad.log1p FALSE NA NA 5.633821e+00 0.000000e+00
## resY.mad.root2 FALSE NA NA 1.669517e+01 0.000000e+00
## resY.mean.nexp FALSE NA NA 1.328912e-110 7.124576e-218
## resY.min.nexp FALSE NA NA 2.543666e-13 7.124576e-218
## max.outdoor.fctr.N max.outdoor.fctr.Y min.outdoor.fctr.N
## .pos 2.000000e+03 1.997000e+03 2.000000e+00
## nImgs 2.974000e+03 1.954000e+03 2.000000e+00
## nImgs.log1p 7.997999e+00 7.578145e+00 1.098612e+00
## nImgs.nexp 1.353353e-01 1.353353e-01 0.000000e+00
## nImgs.root2 5.453439e+01 4.420407e+01 1.414214e+00
## resX.mean.nexp 5.762208e-124 1.209672e-151 7.124576e-218
## resX.min.nexp 3.221340e-27 2.937482e-30 7.124576e-218
## resXY.mad 1.237971e+05 1.078962e+05 0.000000e+00
## resXY.mad.log1p 1.172641e+01 1.158893e+01 0.000000e+00
## resXY.mad.root2 3.518481e+02 3.284756e+02 0.000000e+00
## resXY.mean 2.175740e+05 2.182778e+05 8.762615e+04
## resXY.mean.log1p 1.229030e+01 1.229353e+01 1.138085e+01
## resXY.mean.root2 4.664483e+02 4.672021e+02 2.960172e+02
## resXY.min 1.875000e+05 1.875000e+05 5.000000e+03
## resXY.min.log1p 1.214154e+01 1.214154e+01 8.517393e+00
## resXY.min.root2 4.330127e+02 4.330127e+02 7.071068e+01
## resY.mad 1.942206e+02 1.853250e+02 0.000000e+00
## resY.mad.log1p 5.274130e+00 5.227492e+00 0.000000e+00
## resY.mad.root2 1.393631e+01 1.361341e+01 0.000000e+00
## resY.mean.nexp 1.328912e-110 2.530221e-127 7.667025e-212
## resY.min.nexp 2.543666e-13 7.781132e-20 1.379016e-163
## min.outdoor.fctr.Y max.outdoor.fctr.All.X..rcv.glmnet.Y
## .pos 1.000000e+00 1.999000e+03
## nImgs 2.000000e+00 1.750000e+03
## nImgs.log1p 1.098612e+00 7.467942e+00
## nImgs.nexp 0.000000e+00 1.353353e-01
## nImgs.root2 1.414214e+00 4.183300e+01
## resX.mean.nexp 7.124576e-218 5.762208e-124
## resX.min.nexp 7.124576e-218 2.937482e-30
## resXY.mad 0.000000e+00 1.059169e+05
## resXY.mad.log1p 0.000000e+00 1.157042e+01
## resXY.mad.root2 0.000000e+00 3.254488e+02
## resXY.mean 1.137250e+05 2.182778e+05
## resXY.mean.log1p 1.164155e+01 1.229353e+01
## resXY.mean.root2 3.372314e+02 4.672021e+02
## resXY.min 3.675000e+03 1.875000e+05
## resXY.min.log1p 8.209580e+00 1.214154e+01
## resXY.min.root2 6.062178e+01 4.330127e+02
## resY.mad 0.000000e+00 1.942206e+02
## resY.mad.log1p 0.000000e+00 5.274130e+00
## resY.mad.root2 0.000000e+00 1.393631e+01
## resY.mean.nexp 7.124576e-218 1.328912e-110
## resY.min.nexp 7.124576e-218 2.543666e-13
## min.outdoor.fctr.All.X..rcv.glmnet.Y
## .pos 1.000000e+00
## nImgs 2.000000e+00
## nImgs.log1p 1.098612e+00
## nImgs.nexp 0.000000e+00
## nImgs.root2 1.414214e+00
## resX.mean.nexp 7.124576e-218
## resX.min.nexp 7.124576e-218
## resXY.mad 0.000000e+00
## resXY.mad.log1p 0.000000e+00
## resXY.mad.root2 0.000000e+00
## resXY.mean 8.762615e+04
## resXY.mean.log1p 1.138085e+01
## resXY.mean.root2 2.960172e+02
## resXY.min 3.675000e+03
## resXY.min.log1p 8.209580e+00
## resXY.min.root2 6.062178e+01
## resY.mad 0.000000e+00
## resY.mad.log1p 0.000000e+00
## resY.mad.root2 0.000000e+00
## resY.mean.nexp 7.124576e-218
## resY.min.nexp 7.124576e-218
## max.outdoor.fctr.Final..rcv.glmnet.Y
## .pos 1.200000e+04
## nImgs 2.825000e+03
## nImgs.log1p 7.946618e+00
## nImgs.nexp 3.678794e-01
## nImgs.root2 5.315073e+01
## resX.mean.nexp 5.719134e-133
## resX.min.nexp 9.602680e-24
## resXY.mad 1.086375e+05
## resXY.mad.log1p 1.159578e+01
## resXY.mad.root2 3.296021e+02
## resXY.mean 2.500000e+05
## resXY.mean.log1p 1.242922e+01
## resXY.mean.root2 5.000000e+02
## resXY.min 2.500000e+05
## resXY.min.log1p 1.242922e+01
## resXY.min.root2 5.000000e+02
## resY.mad 2.787288e+02
## resY.mad.log1p 5.633821e+00
## resY.mad.root2 1.669517e+01
## resY.mean.nexp 1.105028e-116
## resY.min.nexp 2.543666e-13
## min.outdoor.fctr.Final..rcv.glmnet.Y
## .pos 2.001000e+03
## nImgs 1.000000e+00
## nImgs.log1p 6.931472e-01
## nImgs.nexp 0.000000e+00
## nImgs.root2 1.000000e+00
## resX.mean.nexp 7.124576e-218
## resX.min.nexp 7.124576e-218
## resXY.mad 0.000000e+00
## resXY.mad.log1p 0.000000e+00
## resXY.mad.root2 0.000000e+00
## resXY.mean 1.058460e+05
## resXY.mean.log1p 1.156975e+01
## resXY.mean.root2 3.253398e+02
## resXY.min 1.802000e+03
## resXY.min.log1p 7.497207e+00
## resXY.min.root2 4.244997e+01
## resY.mad 0.000000e+00
## resY.mad.log1p 0.000000e+00
## resY.mad.root2 0.000000e+00
## resY.mean.nexp 7.124576e-218
## resY.min.nexp 7.124576e-218
## [1] "newobs total range outliers: 10000"## [1] TRUE## [1] "ObsNew output class tables:"
## lunch.-1 dinner.-1 reserve.2 outdoor.3 expensive.-1
## 10000 10000 10000 10000 10000
## liquor.5 table.6 classy.-1 kids.8
## 10000 10000 10000 10000## [1] 0## [1] "glb_sel_mdl_id: All.X##rcv#glmnet"## [1] "glb_fin_mdl_id: Final##rcv#glmnet"## [1] "Cross Validation issues:"
## MFO###myMFO_classfr Random###myrandom_classfr
## 0 0
## Max.cor.Y.rcv.1X1###glmnet
## 0## max.Accuracy.OOB max.AUCROCR.OOB
## Interact.High.cor.Y##rcv#glmnet 0.507014 0.5140745
## Max.cor.Y.rcv.1X1###glmnet 0.502004 0.5290506
## All.X##rcv#glmnet 0.502004 0.5204922
## All.X##rcv#glm 0.502004 0.5202713
## Low.cor.X##rcv#glmnet 0.502004 0.5012952
## MFO###myMFO_classfr 0.502004 0.5000000
## Max.cor.Y##rcv#rpart 0.502004 0.4985602
## Random###myrandom_classfr 0.502004 0.4969618
## Final##rcv#glmnet NA NA
## max.AUCpROC.OOB max.Accuracy.fit
## Interact.High.cor.Y##rcv#glmnet 0.5040041 0.5448976
## Max.cor.Y.rcv.1X1###glmnet 0.5300104 0.5009980
## All.X##rcv#glmnet 0.5108937 0.5299353
## All.X##rcv#glm 0.5238678 0.5236096
## Low.cor.X##rcv#glmnet 0.4978875 0.5265996
## MFO###myMFO_classfr 0.5000000 0.5009980
## Max.cor.Y##rcv#rpart 0.5039318 0.5255967
## Random###myrandom_classfr 0.5059679 0.5009980
## Final##rcv#glmnet NA 0.5109885
## opt.prob.threshold.fit
## Interact.High.cor.Y##rcv#glmnet 0.3
## Max.cor.Y.rcv.1X1###glmnet 0.4
## All.X##rcv#glmnet 0.3
## All.X##rcv#glm 0.3
## Low.cor.X##rcv#glmnet 0.3
## MFO###myMFO_classfr 0.4
## Max.cor.Y##rcv#rpart 0.3
## Random###myrandom_classfr 0.4
## Final##rcv#glmnet 0.4
## opt.prob.threshold.OOB
## Interact.High.cor.Y##rcv#glmnet 0.2
## Max.cor.Y.rcv.1X1###glmnet 0.4
## All.X##rcv#glmnet 0.0
## All.X##rcv#glm 0.0
## Low.cor.X##rcv#glmnet 0.0
## MFO###myMFO_classfr 0.4
## Max.cor.Y##rcv#rpart 0.2
## Random###myrandom_classfr 0.4
## Final##rcv#glmnet NA## [1] "All.X##rcv#glmnet OOB confusion matrix & accuracy: "
## Prediction
## Reference N Y
## N 0 497
## Y 0 501## err.abs.fit.sum err.abs.OOB.sum err.abs.trn.sum
## (32,60] 134.8544 122.3399 258.8735
## (60,120] 123.8999 129.2920 258.0051
## (0,32] 109.1638 117.2849 231.6803
## (120,3e+03] 108.0508 126.8451 239.6148
## err.abs.new.sum .freqRatio.Fit .freqRatio.OOB .freqRatio.Tst
## (32,60] NA 0.2774451 0.2434870 0.2512
## (60,120] NA 0.2564870 0.2605210 0.2459
## (0,32] NA 0.2375250 0.2374749 0.2532
## (120,3e+03] NA 0.2285429 0.2585170 0.2497
## .n.Fit .n.New.Y .n.OOB .n.Trn.N .n.Trn.Y .n.Tst .n.fit .n.new
## (32,60] 278 2512 243 250 271 2512 278 2512
## (60,120] 257 2459 260 251 266 2459 257 2459
## (0,32] 238 2532 237 267 208 2532 238 2532
## (120,3e+03] 229 2497 258 229 258 2497 229 2497
## .n.trn err.abs.OOB.mean err.abs.fit.mean err.abs.new.mean
## (32,60] 521 0.5034566 0.4850879 NA
## (60,120] 517 0.4972768 0.4821008 NA
## (0,32] 475 0.4948730 0.4586716 NA
## (120,3e+03] 487 0.4916477 0.4718376 NA
## err.abs.trn.mean
## (32,60] 0.4968782
## (60,120] 0.4990428
## (0,32] 0.4877480
## (120,3e+03] 0.4920221
## err.abs.fit.sum err.abs.OOB.sum err.abs.trn.sum err.abs.new.sum
## 475.969008 495.761912 988.173744 NA
## .freqRatio.Fit .freqRatio.OOB .freqRatio.Tst .n.Fit
## 1.000000 1.000000 1.000000 1002.000000
## .n.New.Y .n.OOB .n.Trn.N .n.Trn.Y
## 10000.000000 998.000000 997.000000 1003.000000
## .n.Tst .n.fit .n.new .n.trn
## 10000.000000 1002.000000 10000.000000 2000.000000
## err.abs.OOB.mean err.abs.fit.mean err.abs.new.mean err.abs.trn.mean
## 1.987254 1.897698 NA 1.975691## [1] "Features Importance for selected models:"## All.X..rcv.glmnet.imp Final..rcv.glmnet.imp
## resX.mean.nexp 100 100
## resY.mean.nexp 100 100## [1] "glbObsNew prediction stats:"##
## N Y
## 0 10000## label step_major step_minor label_minor bgn end
## 22 predict.data.new 10 0 0 261.304 277.318
## 23 display.session.info 11 0 0 277.319 NA
## elapsed
## 22 16.015
## 23 NANull Hypothesis (\(\sf{H_{0}}\)): mpg is not impacted by am_fctr.
The variance by am_fctr appears to be independent. #{r q1, cache=FALSE} # print(t.test(subset(cars_df, am_fctr == "automatic")$mpg, # subset(cars_df, am_fctr == "manual")$mpg, # var.equal=FALSE)$conf) # We reject the null hypothesis i.e. we have evidence to conclude that am_fctr impacts mpg (95% confidence). Manual transmission is better for miles per gallon versus automatic transmission.
## label step_major step_minor label_minor bgn
## 1 import.data 1 0 0 24.223
## 16 fit.models 8 0 0 132.969
## 2 inspect.data 2 0 0 79.636
## 17 fit.models 8 1 1 186.824
## 20 fit.data.training 9 0 0 229.013
## 22 predict.data.new 10 0 0 261.304
## 18 fit.models 8 2 2 213.842
## 3 scrub.data 2 1 1 116.110
## 21 fit.data.training 9 1 1 254.865
## 19 fit.models 8 3 3 224.757
## 15 select.features 7 0 0 129.924
## 14 partition.data.training 6 0 0 128.635
## 11 extract.features.end 3 6 6 127.205
## 12 manage.missing.data 4 0 0 128.135
## 13 cluster.data 5 0 0 128.567
## 9 extract.features.text 3 4 4 127.083
## 6 extract.features.datetime 3 1 1 126.926
## 7 extract.features.image 3 2 2 126.988
## 10 extract.features.string 3 5 5 127.150
## 4 transform.data 2 2 2 126.865
## 8 extract.features.price 3 3 3 127.049
## 5 extract.features 3 0 0 126.905
## end elapsed duration
## 1 79.635 55.412 55.412
## 16 186.823 53.854 53.854
## 2 116.109 36.474 36.473
## 17 213.841 27.017 27.017
## 20 254.864 25.851 25.851
## 22 277.318 16.015 16.014
## 18 224.756 10.914 10.914
## 3 126.864 10.754 10.754
## 21 261.304 6.439 6.439
## 19 229.013 4.256 4.256
## 15 132.968 3.045 3.044
## 14 129.923 1.289 1.288
## 11 128.134 0.929 0.929
## 12 128.566 0.431 0.431
## 13 128.635 0.068 0.068
## 9 127.149 0.066 0.066
## 6 126.987 0.061 0.061
## 7 127.048 0.060 0.060
## 10 127.204 0.054 0.054
## 4 126.905 0.040 0.040
## 8 127.083 0.034 0.034
## 5 126.926 0.021 0.021
## [1] "Total Elapsed Time: 277.318 secs"